Agenda for HOLLIS Liaisons Meeting #89
October 13, 1993
Lamont Forum Room, 9:30 - 11:00 a.m.
1. Announcements: Jon Rothman
2. Automation Planning Committee, status report: Tracey Robinson
3. Recon authority processing update: Robin Wendler In mid-October, with authority processing on the HU base file complete, Harvard will begin sending cataloging and receiving corrections from OCLC on an ongoing basis. Robin will review exactly what will happen when, and answer any questions about the ongoing heading correction process.
4. Review of HOLLIS indexing enhancements: Jon Rothman Liaisons will review the indexing section of the enhancements list. OIS invites liaisons and other interested staff to come ready to suggest additional enhancements and choose the top five. Jon will also review a separate list of indexing enhancements that are part of the LISH report.
Notes from the September Meeting
OIT mainframe upgrade. OIT has scheduled a memory upgrade on the weekend of September 18 to improve system capacity and response time problems. All systems will come down on Saturday the 18th at 6:00 pm for 12 to 30 hours. All systems should be available again by Monday morning and perhaps as early as mid-day on Sunday. [From the Editor: the upgrade went smoothly, although there were unrelated network and ISDN communications problems that weekend that may have delayed availability of HULPR/HOLLIS for some staff. These problems have been resolved.]
Upcoming meetings of interest. Tracey reminded liaisons of several upcoming meetings, noted in the Calendar section of the September newsletter. These include the Cataloging Discussion Group on the 10th, Tech Services Managers Group on the 15th, and the first meeting of the Library Microcomputer Interest Group on October 13th.
Follow-up on 050 field discussion. As a follow-up to the August discussion on keeping or deleting 050 fields with textual information, Robin Wendler asked liaisons to consider one other issue related to the 050 and 090 fields during HOLLIS cataloging processing. When incoming utility cataloging is processed, if a HOLLIS record contains an 050, any 090 fields are dropped during duplicate resolution. Those units claiming records for local editing in HOLLIS are not having the call number moved into their LOC automatically, so when the 090 is dropped, the only usable call number in the record may be lost. The group agreed that duplicate resolution in HOLLIS should be changed so that all unique 050 and 090 fields are retained. In addition, the group agreed that the LCCLAS field should not reflect the non-call number 050 information. [From the Editor: upon further review, OIS has determined that when a valid LC number is not present in the 050, the LCCLAS field is left blank. The necessary change therefore, is simply to make sure that if the HOLLIS record has a blank LCCLAS and the incoming record contains LCCLAS data, that the data is retained when the records are merged.] Contact Robin Wendler in OIS if you have questions.
ERIC database comes to HOLLIS. The new ERIC database will be quite similar to the existing HOLLIS databases, but with two special characteristics. There is a new index identifier, KSM (keyword subject major), which is a subset of the KSH index, and contains only topics significantly covered in an indexed article or document. This index is only available in the ERIC database because ERIC is the only data supplier which makes this distinction. The other special aspect of ERIC is the way the LOC command works, or rather, where it doesn't work. ERIC indexes both whole documents and journal articles. Many of the indexed journals are available on campus and you can use the LOC command to determine which libraries hold a particular journal title. However, nearly all of the documents are held at Harvard only in the Gutman Library ERIC microfiche collection. The ERIC microfiche number, which serves as the Gutman call number, appears in the NOTES area of each ERIC document record. If someone issues the LOC command from an ERIC document record, instead of sending them on a fruitless search of HU, the message "Check for Harvard location in NOTES. Type HELP for details." will display. Contact Robin Wendler or Julie Wetherill in OIS if you have questions about ERIC in HOLLIS.
Overview of Course Reserves project. Kate Weber gave an overview of the Course Reserves project. GARP (the Group for Automating Reserves Processing) convened last year and produced a report that was approved by HAAC this summer. GARP recommended a two-pronged approach: on the mainframe, a new course reserves database in HOLLIS that would be separate from HU and contain reserves information, locations, and linked item records with circulation status. This database would have its own indexes and would provide a catalog mode display of course reserves information. The second prong of the report recommends using the power of microcomputer-based wordprocessors, spreadsheets, and database management software to manage the many simple, repetitive, and time- consuming tasks that are part of reserves processing and reporting. To facilitate the microcomputer portion of this process, GARP has recommended expanding the HOLLIS/HULPR downloading process. A subgroup of GARP is investigating options for this process.
Kate noted that the functional and technical specifications for the mainframe portion of this project are now being written. Daniel Bednarek is consulting with members of GARP to define the additional reporting functionality needed to support the microcomputer portion of the project (see related note about the records-by-email function). SSHUSH (Standing Subcommittee for User Services in HOLLIS) is reviewing some public catalog issues. SSHUSH and GARP members will develop the necessary catalog help screens. Library reserve departments will be defining standards for input into the course reserves database and getting a handle on the necessary microcomputer expertise. This last group will be working over a period of 1-2 years. OIS will keep everyone informed through this newsletter as development proceeds. Please contact Kate Weber in OIS if you have questions about this project.
OIT's Network Information Center in full gear this Fall. Annelise Martin, OIT's Manager of User Services, provided an overview of OIT's new Network Information Center (NIC). Since the 1960's OIT has maintained a User Support Desk for mainframe systems, then microcomputer operations, and more lately networking. Staff have been answering HOLLIS technical questions since 1989. OIT has now consolidated its various user support numbers into one Network Information Center number (496-2001, Monday-Friday 9:00 am-5:00 pm). The NIC currently receives approximately 100-120 calls per day at peak times (Sept/Oct. and Jan./Feb.). NIC staff are working closely with various faculties to insure that students are familiar with and have access to the NIC support structure. Through the Fall semester, users can call any of the previously published support numbers and will be forwarded to the NIC. Later this year, calling an "old" support number (such as 495-9388) will produce a message telling you to call 496- 2001. (Note: NIC staff are not the Help Desk staff at 495-3000. Continue to call 5-3000 for production system problems.)
There is a NIC brochure available from the Harvard University Copy Center. Just call and give your billing number and the number of copies desired. In addition, OIS will distribute small supplies of the brochure this fall. Naomi Ronen asked that OIS remember to change HOLLIS help screens and information sheets to reflect the new NIC telephone number.
So, call 496-2001 with your microcomputer or network questions. Consult an article in the August HOLLIS Newsletter that describes some of the other services offered by the NIC.
Recon update. Karen Carlson Young reviewed the latest Recon happenings. OCLC has reached the next to last billing period for the first year of Recon and they anticipate that they will have met or exceeded production goals by year's end. The Kummel project is now complete and Cabot is more than halfway done. Wolbach will be the next monograph project to start. The hit rate (records found and claimed) varies widely from project to project, from a low of 54% at Design to high of 86% at Kummel (the higher the hit rate, the faster the processing).
The Serials Recon project is starting slowly with four boxes of MCZ serials searched in HOLLIS and sent to OCLC. There was a 15% hit rate in HOLLIS for these serials (within project expectations), largely on CRL records. The remaining 700,000-800,000 correction transactions from processing of the HU base file are due back in the next few weeks and will be applied to the HU database by mid-October. The original estimate was closer to two million transactions, but OCLC discovered that the HU database was "cleaner" than originally anticipated.
Update on the Bridge project. Jon Rothman updated liaisons on development of a network-based menu system -- the Bridge -- that would provide staff and patrons access to local and remote databases, such as HOLLIS, VINE, RLIN's CitaDel, and OCLC FirstSearch files. A HAAC task group, unfortunately named GIRLS (Gateway to Information Resources and Library Systems), produced the initial recommendations and a later group, "Son of Girls," is now finalizing specifications for the displays, commands, etc. -- their report is due soon. OIS is continuing technical work and hopes to have a beta test version for demonstration by end of this calendar year. Tracey added that much of the delay in Bridge development can be attributed to the technical limitations presented by mainframe- connected terminals (i.e. our public terminals). There is no easy solution, and we console ourselves by accepting that the world is moving toward microcomputer access. The character set is also a sticking point -- units may have to choose between Bridge access without the character set and dedicated HOLLIS access with the character set. Tracey encouraged staff to keep watching and listening for news about the Bridge.
Spine label update- (may this please be the last one!). Jon Rothman lost the draw and was elected to present the update on the spine label function in HOLLIS. In August, beta-testers turned up three bugs in the Vernon spine label software. OIS has reported these to the manufacturer and hopes to have a fix in time for mid-September. When distribution time is near, Julie Wetherill will contact units that have expressed interest in this feature to finalize distribution of software and documentation.
Update on the End-User Reporting project. This project aims to put more HOLLIS reporting in the hands of library staff and possibly increase the range of reports OIS can offer centrally. Jon Rothman reported that OIS still considers OIT's Information Utility (IU) to be the best prospect for storing the reporting data. The IU is a mainframe database containing personnel, financial, and development data. OIS would use the IU for mediated reporting; staff could produce "canned" reports from the IU; and knowledgeable users could query the IU directly (using SQL) to construct their own reports. Reporting output could be in printed or electronic form. A working group of interested users is defining the set of reporting data elements. There is no firm date, but OIS hopes for beta testing to begin early in 1994. Contact Jon Rothman in OIS if you have questions.
Course Catalog update. The Harvard Course Catalogs are up and course schedules for all Faculties are available. Tracey Robinson noted that at present there is network and dial-up access but no dedicated HOLLIS terminal access to Course Catalog data (see the September Newsletter for more information). Contact Amy Lozano (495- 1821) for more information.
Business Resumption update. The OIT Business Resumption Project has been underway for several years. The goal of the Project was to develop a strategy and plan for coping with a major disaster at the computing center causing an extended outage of the mainframe computer on which HOLLIS and HULPR reside. Tracey Robinson reported that June marked a significant event in the project in the successful completion of a disaster recovery test on a commercial "hot site" computer in Philadelphia. The test revealed that it took approximately 12 hours for OIT staff to recover the mainframe operating system and basic system utilities. It took OIS staff an additional 12 hours to recover the HOLLIS and HULPR systems. Recovery of HOLLIS includes bringing HOLLIS/HULPR files back to the state they were in as of the last system backup procedure. Depending on the exact timing of a "disaster" we are currently in a position to recover HOLLIS/HULPR to the state they were in 24-30 hours before the disaster occurred. (Backup tapes are taken offsite at 1:00pm daily. So, for example, if the disaster occurred at 11:00pm on Wednesday, we would lose all activity performed on Wednesday. Files could be restored which reflect activity through Tuesday. If the disaster occurred at 11:00am on Thursday, we would still only be able to restore through Tuesday because the Wednesday backups would not yet have been taken offsite.) Library staff should be aware of this "window of vulnerability;" that is, they might be required to manually reconstruct up to 1-1/2 days worth of online work.
The other vulnerability is the connection to Philadelphia -- if your staff and public HOLLIS devices are on the HSDN there would be no problem in accessing HOLLIS/HULPR on a remote computer (in Philadelphia or elsewhere). If you depend on hard-wired terminals attached to muxes and telephone lines, you would be out of luck until phone line connections to the remote site of HOLLIS could be established (this could take several weeks). It is possible you could dial the remote site with a modem. Another part of business resumption -- an intermediate backup for a mid-size disaster (using the DUC, CD-ROM, RLIN, etc.) -- is still under discussion. Contact Tracey Robinson in OIS if you have questions.
Notes and Reminders
Change in patron file update schedule. OIS and the Office for Human Resources are making the final arrangements to increase the frequency of patron file updates to HOLLIS from monthly to weekly. When a firm schedule is set OIS will announce the details in this newsletter and via HULINFO. Contact Daniel Bednarek in OIS if you have questions.
Technology calendar solicitation. The HUL Professional Development Committee is once again this year planning to compile a monthly calendar of events related to the topic of "technology and libraries". The calendar will be posted to LIBINFO, as well as distributed in printed form (maybe we will go wild and use a different color this year!!). This is a request to all members of the Harvard library community. I ask that you assist me in this endeavor by forwarding information to me on events, programs, etc. that are likely candidates for inclusion in the calendar -- mostly I am looking for "local" events that are accessible for Harvard library staff. I would appreciate your help.(TR)
Request for info on use of record-by-email function. As part of the project to mount an automated course reserves database in HOLLIS, an investigation is under way to examine methods for extracting, or downloading, HOLLIS records for later manipulation by database management software in a PC or other computing environment. This will allow users of the course reserve system to control locally any unique data management or report generation needs.
This is a request for any examples of (or information about) methods users have developed for loading data that was downloaded using the HOLLIS Public Catalog records-by-email function (a.k.a., STORE/SEND), style B. Specifically, I am looking for answers to the following questions:
Any information would be useful, including names of people who have been using the STORE/SEND feature of HOLLIS in this way. Please call Daniel Bednarek at OIS or mail to BEDNRK@HARVARDA.HARVARD.EDU if you have questions or information.
Change in ACU report frequency. OIS has received a request to alter the frequency of the Accounting Unit Status (ACU) reports. These reports are currently produced weekly on Sunday night for distribution with Monday products. The request is to alter this to the 15th and last day of each month (semi-monthly) for July 1 to February 28 but to maintain the weekly schedule from March 1 to June 30. Since this proposal went out on HULINFO there have been no objections. OIS will go ahead and run the ACU reports on September 30th and this new semi-monthly schedule will remain in effect through February 1994, after which frequency will change to weekly through June. Contact Linda Marean in OIS if you have any questions.
New implementation date for format integration. This article comes courtesy of Winston Tabb, Associate Librarian for Collections Services, Library of Congress. Robin Wendler posted this message to LIBINFO earlier in September.
In June 1993 the Library of Congress announced a delay in its implementation of format integration, with implementation to occur in late 1995 rather than in early 1994, as LC had originally projected. Following this announcement the Research Libraries Group (RLG) proposed that implementation be divided into two phases, with the first phase being completion of the additions and changes to the variable fields, tags 010-8XX, by the end of 1994 and the changes to the Leader, 006, 007, and 008 fields by late 1995.
To determine the most satisfactory national approach, the Library initiated internal and external consultations with the utilities and with participants on the USMARC electronic forum. There was general support for the RLG proposal, and the networks felt the division of the job into two phases was useful for them. The new target dates are possible for LC, and the division of labor is deemed highly desirable by our systems staff. The emphasis in all comments was that the schedule be met!
I will do my best to assure that LC meets these targets. The Network Development and MARC Standards Office will issue the new edition of the bibliographic format in early 1994. This new edition will contain the integrated 010-8XX fields.
We appreciate the cooperation we have had from our colleagues in working through this difficult problem. We all look forward to completion of this task and to the many benefits that format integration will bring us. (Winston Tabb) [From the Editor: if you have questions about this schedule or about its impact at Harvard, contact Robin Wendler in OIS.]
Changing liaisons. Danila Terpanjian is the new HOLLIS liaison representing Littauer Library. Danila replaces Steve McGinty, who has left to pursue his Doctorate. If you regularly contact the Littauer liaison via e-mail, you should use Danila's address: DANILA@HARVARDA.
Recon Central staff began searching the MCZ serials in late July and completed the searching in early September. The cards for which the MCZ location and holdings were added to HOLLIS were marked with a red dot so that OCLC operators can easily identify them as cards to skip in their searching, since these titles are considered converted. Once at OCLC the operators search and claim matching records and key records for titles not found in the database. Editing is done according to a set of instructions prepared by the Recon Project Manager and library staff.
Serials converted through the Retrospective Conversion Project can be identified in several ways:
See HOLLIS record number AQN2757 and associated holdings for an example of a converted serial title. Gutman is next in the serials queue, and it is expected that the first batch of serials will be sent to OCLC by the end of September.
Response Time Testing. OIS has decided to discontinue monthly response time tests of HOLLIS and HULPR until further notice. Due to the fact that we have recently "lost" several of our response time volunteers, and coupled with the growing complexity of network connections, OIS is investigating how to collect response time data in the future. Volunteers should return their stop watches to OIS. Contact Julie Wetherill in OIS if you have questions.
HOLLIS Enhancements Update
1. The training file (TR) is now available in the OPAC (with safeguards to prevent unwary users from confusion, as recommended by liaisons). It is now possible to try changes in TR staff mode and see their effect in TR catalog mode. To try this type "TR" from the database selection screen or "CHOOSE TR" from anywhere in HOLLIS.
2. A new parameter to the SET TAP command has been implemented so that it is possible to be certain that there is not a current tape out destination set. The new command is issued in the form: "LTHU SET TAP OFF".
3. A method was devised for handling flat fee circulation overdue fines and producing notices with the appropriate text. The text on overdue recall notices for Widener was changed to "RECALL fine is $10."
4. Changes were made to the types of terminals defined to HOLLIS to accommodate access by devices running the McGill TCP3270 package. This package is an alternative to the Cornell PCIP software, which allows telnet to HULPR with the full ALA character set.
In addition to the changes listed above, 2 additional small changes which affect the HOLLIS infrastructure were completed recently. "Infrastructure changes" are enhancements and bug fixes that are needed to ensure reliable and efficient operation of HOLLIS but which do not have directly visible or readily describable functional effects.
Where authority cross references do not belong. This is just a reminder that if you want to provide access points to multiple versions of a name, you should rely on authority record cross references rather than adding the names to a bibliographic record. As an illustration, the following biographic record contains multiple added entry fields in an attempt to provide access to variations in the form of the name National Oceanic and Atmospheric Administration:
The cross references in the 410 (series entry) fields belong on an authority record. The proper form of the name is correctly recorded in the second 710 field on the bibliographic record and it also is established in a separate authority record, which includes name variations (below):
Please note that the real bibliographic record in HOLLIS (AMZ9104) is currently correct. The Editor "doctored" the example here only to illustrate the point. If you ever have a question about authority work, consult a cataloger or contact Robin Wendler in OIS.
Using DERIVE with authority records. Some months back, Jeff Beall posted a suggestion to LIBINFO about using the DERIVE command with authority records. This message is repeated below:
Two types of authority records where the DERIVE command often can be used are uniform title authority records (UTI HAFs) and for series authority records (SARs), especially when inputting a subseries SAR when a SAR for the main series already exists.
For example, if you wanted to create a UTI HAF for a new translation of a work by Gabriel Garcia Marquez, you could pull up the existing HAF for his One hundred years of solitude, type DERI and press ENTER to derive a new authority record, and then modify this new record to correspond to the book you are cataloging (i.e., change the subfield t's in the 100 and 400 fields). The encoding level always defaults to "c," so you routinely have to change this fixed field value to "h." (Jeff Beall)
When do automatic claims get produced? This is one of the most frequently asked questions in HOLLIS acquisitions. An "automatic claim" is the memorandum that HOLLIS produces automatically for unreceived one-time orders and for serial orders when the first piece is unreceived and overdue. On the o/p/r, the evidence that an automatic claim has been produced will be a memo statement similar in form to the following:
First microcomputer users meeting in October. The first meeting of the library microcomputer users group will take place October, 2:30-4:00pm in the Lamont Forum. The purpose of the meeting will be to explore how future meetings of the group will be organized and what general areas of microcomputer use in libraries will be addressed by the group. Examples of the latter might include: the development of an integrated cataloger's workstation, reporting (statistical applications), Internet applications like Gopher and WWW, etc. Hopefully, agenda topics for the next few meetings of the group will be decided upon. All are welcome to attend. For more information, contact Heather Reid in OIS or by e-mail (firstname.lastname@example.org).
Most of the data stored in HOLLIS is considered to be "public" information. Exceptions to this rule include accounting information and some circulation information such as patron and billing data. Restricting access to patron information is of special concern to librarians because we feel a professional responsibility to protect the privacy of library patrons. This statement describes the two basic types of patron information available in HOLLIS and explains the privacy issues related to each.
One type of patron information identifies or describes a patron himself, for example, social security number, address, or phone number. A second type lists or identifies items that a patron has checked out, currently or in the past. The number of operators who have the authority to view these types of information is intentionally limited to as few as possible. For those operators who do require access to patron information to perform their daily work, this statement is intended to provide an explanation of the privacy issues and some guidelines to be followed in order that all library staff share in the responsibility to protect the privacy of library patrons.
The HOLLIS patron file is a valuable tool for libraries due to the fact that it reflects data from the central university Human Resources file. It is used in HOLLIS to control circulation, as well as to control access to databases in HOLLIS which have usage restrictions imposed by the vendors that supply Harvard with special data (for example, journal indexes). The information contained in the patron file should be considered "private" and should only be used by library staff to perform normal library functions. Information from a patron record should never be made public or given to anyone other than the patron in question. Addresses, phone numbers, and identification numbers are private information and should be treated as such. The patron file should never be used as a directory. The fact that the patron file has personal data that individuals can have suppressed from public directories, phone books, etc. reinforces the point that this information is private.
The second type of patron information includes any information related to the borrowing details or habits of library patrons. In accordance with the American Library Association policy on the confidentiality of library records, library staff should recognize that "circulation records and other records identifying the names of library users [are] confidential in nature". In other words, information relating to what an individual has checked out (currently or historically) is private information. Positive identification must always be required of a person requesting information concerning materials he or she has on loan. Limited access to borrower information by library circulation staff may be necessary to resolve certain problems pertaining to circulation processing. Under no circumstances should a library staff member reveal to anyone other than the patron in question the identity of a person who has a certain book checked out. Likewise, recall and "on hold" information should be considered to be private information and should never be made public. Privacy pertains equally to specific circulation records (e.g., patron A has book X checked out) as well as to patterns of circulation (e.g., patron A borrows materials in subject area Y). No historical records are kept which would enable us to identify the past circulation activity of any individual.
Special care must be taken to insure that access to patron information and patron circulation activity is used only by authorized library staff for appropriate library functions. Care should be taken to make sure that a terminal with access to patron records is never left logged on and unattended. Further, a terminal placed in a public service area which is used to access patron information should always be positioned so that the terminal screen is not viewable to the public. Needless to say, logon names and passwords should never be shared or be written on the bottom of a keyboard be made accessible to unauthorized people in any other way.
In summary, all library staff share a responsibility to protect the privacy of patron information. If you have any questions about this policy or encounter a situation in which your responsibility for upholding the privacy in unclear, please discuss it with your supervisor.
As part of the HOLLIS Administrative Advisory Committee's policy on access to HOLLIS patron records (January 1990), the Committee recommended that any staff member with authorization to access patron records should be required to read the ALA policy on "Confidentiality of Library Records" and to review said policy on an annual basis. OIS periodically reproduces the ALA policy as part of the HOLLIS Newsletter. Circulation supervisors are encouraged to distribute a copy of the policy to all appropriate library staff.
("Confidentiality of Library Records", from the ALA Handbook of Organization 1989/1990 and Membership Directory, 1989, pp.243-244)
"The American Library Association strongly recommends that the responsible officers of each library, cooperative system, and consortium in the United States:
1) Formally adopt a policy which specifically recognizes its circulation records and other records identifying the names of library users with specific materials to be confidential.
2) Advise all librarians and library employees that such records shall not be made available to any agency of state, federal, or local government except pursuant to such process, order, or subpoena as may be authorized under the authority of, and pursuant to, federal, state, or local law relating to civil, criminal, or administrative discovery procedures or legislative investigatory power.
3) Resist the issuance or enforcement of any such process, order, or subpoena until such time as a proper showing of good cause has been made in a court of competent jurisdiction."