Library Technology Guides

---

Volume 4 Number 11 (November 1984)

Integrated online systems conference

The Integrated Online Systems conference held in Atlanta on October 18 and 19 included a number of presentations of potential interest to LSN readers.

One of the speakers, John Corbin, Assistant Director for Administration and Systems at the University of Houston, addressed some of the personnel and management issues which arise when incorporating a multilocation integrated library system into an existing library environment. The following statements were excerpted from his presentation:

The introduction of an integrated library system brings with it major changes in the relationships among different functional areas of a library. In libraries with no automation, functional areas operate independently; the addition of some automation results in a degree of interaction among functions; the implementation of integrated automation brings with it the total interdependence of functional units. Such integration requires careful planning and coordination. Staff must be aware of the activities of all other work units. Attention to accuracy and workflow is essential with an integrated system because what were previously isolated errors, slow-downs or blocks in the workflow, can now "ripple" throughout the entire library system. Small problems can become quite large.

As an integrated automated library system brings increased interdependence, it also brings greater visibility. The staff will be placed under increased scrutiny by their peers and the public. "Inner sanctums" are breached, and action or inaction become visible to all.

Shifts in workloads will also follow the installation of an integrated automated system. A blending and reorganization will occur, along with a reduction in the paperwork generated. In shared systems serving multiple locations there is a transfer of some management functions, and a need for the development of new management positions for outreach/ linkage among the separate service points.

With a shared system, the increased workload in areas such as interlibrary lending can be dramatic. The system's ability to identify and locate system wide resources, and to determine their availability, may soon lead to demands for a quicker and more efficient document delivery system.

Implementation of a shared automated library system will decrease the local autonomy of campus departments, branch libraries, or multi-campus libraries. Since local records are entered into a common file, the identification of personnel authorized to make changes in records, the determination of standards, and the monitoring of adherence to those standards all become issues on which consensus is required.

One way to organize the management of a multi-campus automated library system is to locate the system manager, the managers for each of the component functions, system operators, and training personnel in a separate facility close to the computer system. A second way is to house these staff in one of the participating libraries, along with the necessary computer support. There are political considerations either way. The second method provides no separate funds for housing the staff and support facilities but gives the host library the possibility of exerting greater control over the system and system managers. Whatever method of staffing is chosen, -it is essential that there by an advisory commission made up of the directors of the participating libraries.

An integrated library system comprises a complex set of equipment, software and people which provides opportunities for cooperation and communication. To fully exploit its opportunities may require a reorganization of the complete library situation from top to bottom.

In conclusion, Corbin stated that the introduction of a multi-campus integrated automated library system at the University of Houston has brought changes in the libraries' requirements for entry-level staff and has increased the need for continuing education for existing staff. There is now much more emphasis on computer literacy for library positions.

N. A. Huttner, Deputy Director of the Cleveland Public Library, addressed a similar situation in recounting that library's experience in implementing an online system in a multi-library setting.

In 1976 the library began a search for a circulation system and a data base. Data Research Associates, Inc., of St. Louis was chosen to provide both. In 1980 online patron access catalog capability was added. However, the card catalog remained in place. By December 1982, the data base of some 2 million records was loaded. All 31 branches and the main library were online for circulation and patron access catalog by December 1982.

Cleveland Public Library (CPL) is a major research library with 1.2 million titles and 7+ million tags searchable or the system. The question was, how a library such as CPL could share an integrated system with a suburban library system. Following a tentative cost analysis, a contract was devised for cooperation within a network system with the parent library providing advice and training to new members.

The resulting network handles some 5 million circulations a year. In addition to the more obvious benefits, automation has also promoted cooperative weeding and collection management. However, it has been difficult to ensure that member libraries retain their individualities and continue to serve their clients' unique needs. Is it necessary that all members follow the same uniform rules? Cost often dictates the solutions while organized standards remain the objective. The thrust toward cooperation continues.

Among the questions that must be faced in such networks are: who owns the data and the data base; what happens to libraries who want to leave the system; and how can member libraries be represented in the OCLC data base if they don't use OCLC cataloging? How large can a regional system be? What are the practical geographical limits? Huttner concluded the presentation by emphasizing the importance of finding answers to questions of how individual library autonomy can be retained in an atmosphere of reasonable cooperation, and how library staff can continue to provide personalized service to users after the implementation of a cooperative automated system.

Stephen H. Salmon, President of Carlyle Systems, Inc., Berkeley, California, discussed how new and emerging approaches to integrated online library systems such as data processing and remote time sharing are changing the scope and definition of an integrated online system. It may soon be feasible to combine selected modules from integrated systems with a single function system to configure a unique package tailored to a library's particular needs. These separate modules could then be interfaced to function as a single integrated automated library system.

Salmon described an integrated system as one in which two or more modules work off the same data base. The functions might include: acquisitions, cataloging, serials control, circulation, online patron access catalog, catalog maintenance, authority control, ILL, union list, reserves, collection development, special collections, spine labels, and newspaper index, as well as management information on all of the above. To expect a system to include the entire list is unrealistic, but current RFP's received by Carlyle Systems have included all of them. Salmon commented that unrealistic expectations on the part of libraries often lead to unrealistic promises by vendors.

Available functional capabilities are only just beginning to meet the needs of libraries in terms of software and hardware development. To date, system hardware has been based on minicomputers. A new approach, being used by Geac among others, is to use multiple processors for discrete operations similar to distributed processing systems. In distributed processing the systems can run independently using appropriate software. For example, one machine handles general functions separately from the data base management functions. The hardware can be optimized for function. The advantages of this type of system are the cost efficiency of standalone independent operation and the ability to run the various functions under different operating systems.

Data base management functions run on a separate machine. The same can be done for other functions using a modular design including an interface processor, a remote processor, and a data base processor. Enhancements are possible because the microprocessors use parallel processing. The micro enhancements, based on single board chips, are capable of performing multiple functions with high utility and lowered costs.

At present, this a la carte or modular approach to configuring an automated library system is a possibility on the horizon rather than a solid option. It is coming, but cannot be achieved until the different system vendors agree to cooperate and work to that end. The approach requires that the terminals of one system serve as terminals to the other systems. The advantage for libraries would be that they would be able to choose from the various vendors, the functional modules which best suited their individual needs.

Online ordering update

Intrigued by conflicting reports of the interest in online ordering being displayed by different types of libraries in the United States, the Editors recently undertook a brief telephone survey to gain some current insight.

In 1982, as part of a study conducted for Library Technology Reports, we had queried 80 academic, public and special libraries about the essential characteristics of automated acquisitions systems. At that time nearly half of the respondents at major public libraries- those with over $1 million a year in acquisitions funds-said that online ordering would be an important feature of any acquisitions system they might implement. At the same time, only 10 percent of the academic libraries considered online ordering an important capability. On the other hand, the special librarians interviewed--almost all of them in the corporate sector-considered online ordering more important than either of the other two groups, with two-thirds choosing it as a mandatory system element.

To determine whether or not these percentages had changed, we first checked with 10 of the libraries that had participated in the 1982 survey. Two had been using an automated acquisitions when contacted before; two more had implemented systems since then. Two public libraries which have automated systems are ordering online; two academic libraries with automated systems are not. The reason the public libraries are ordering online is that it makes it possible for them to get "best sellers" on the shelves in time to satisfy the demand stimulated by advertisements and reviews. In general, the responses from the 10 libraries resurveyed had changed very little from the previous survey.

We also contacted a number of large academic libraries not previously surveyed. Of the 22 institutions surveyed, 14 currently use an automated acquisitions system-10 use in-house systems and 4 the RLIN acquisitions subsystem. Two are installing the multifunction BLIS system and plan to implement its acquisitions module. Only 6 of the in-house systems are online systems and all but 3 of these are single function systems. (All are likely to be replaced in the next five years or so and in only one case is it certain that the new system will also be an in-house development.)

Only one of the libraries is currently using an online ordering system, Faxon Linx, for serials ordering. None of the libraries has experience with on-line ordering of monographs and other firm orders.

Of the libraries not currently using automation in acquisitions, only two definitely plan to do so within the next two years. One of them has a DataPhase integrated library system and will bring up the acquisitions module within the next year or so. The other has not selected a system. Six others believe the time frame for acquisitions automation will be five years or so because the library has other automation priorities, especially implementing an online patron access catalog.

Future online ordering is not a high priority with any of the libraries. Only four consider it likely that they will have it within the foreseeable future.

Decisions to implement online ordering would be dependent on cost and the number of vendors that could accept the transmissions. For ordering, jobbers are generally deemed to be more important than publishers. Several of the interviewees stressed, however, that while online ordering capability would be a selection criterion in procuring an acquisitions system, it would not be a mandatory specification element. The use of online ordering would be substantially dependent on whether the most attractive turnkey system or software package included the capability. Online ordering would also be more attractive if the major jobbers of a library could all accept online orders.

Several respondents said that a decision would have to be tied to benefit. If the online ordering system cost was substantially offset by savings in personnel, mailing, and supplies it would probably be adopted. However, speed alone would not be enough. The majority pointed out that the time saved by ordering online would be lost while the items waited for cataloging. Six respondents said that they now use the telephone and telegraph for rush orders and doubt that they could implement online ordering for the small number of rush orders they process at a cost lower than the aggregate annual cost of using the telephone and telegraph.

Six of the respondents volunteered that if vendors begin to accept online ordering, they should also consider online invoicing. They pointed to the fact that many book stores now have online ordering and invoicing service from their suppliers.

The responses reflect the librarians' current thinking which has not changed appreciably from the 1982 LTR survey. There is evidence to suggest however, that the importance of a feature increases when it becomes widely available. For example, the online printer port interface between OCLC and the CLSI system was deemed important by fewer than 20 percent of a group of librarians interviewed in 1979. Two years later-when printer port interfaces between bibliographic utilities and local library systems had become generally available-over 75 percent deemed them to be essential. The pattern was similar for call number access to circulation records- from low priority in 1978 to essential for over 90 percent of libraries in 1983. Only time will tell if wide availability of online ordering will have a similar effect.

Loading separate bibliographic files onto a Geac system

A piece in the September 1984 issue of LSN describing the file loading and merging capabilities of the CLSI, Data Phase, and LS/2000 automated library system prompted an inquiry about Geac's capabilities in this area. The established Geac procedures allow libraries wishing to merge separate bibliographic files into a single bibliographic data base to nominate a hierarchy of selection criteria to be applied to the selection of the record to be retained when the files to be merged contain more than one entry for a single bibliographic item. Applicable criteria include cataloging source and library or branch symbol.

Once the selected procedures are applied, the system retains all of the record that is to be the master record and adds to it the local copy and call number information from the records it is to replace. Other local data such as local subject headings or added entries are not retained.

The merging operation is usually carried out at the library. The tape drive and merging programs used are part of the regular Geac system configuration and, after the initial merge and load, are used to support any required ongoing loading of records from tape. There is a one-time charge of $5,000 for setting up the merger parameters if the files being merged use the MARC format. When the process involves the handling of non-MARC records, there is an additional charge for the necessary custom programming. A spokesperson indicated that this charge rarely exceeds an additional $5,000.

Reflective optical videodisc applications

As the popular press continues to highlight new and exciting applications of various optical" media, many librarians have expressed concern as to the stability and viability of the subset of these technologies of most immediate concern for both collection development and emerging library data storage applications-videodisc.

Currently, only two types of videodisc are receiving commercial support in the U.S.-the grooved capacitance CED disc adopted by RCA and the laser or reflective optical disc sold by Pioneer and Sony. Although RCA continues to abide by its promise to continue to produce new entertainment programming for its CED disc players for several more years, the continued viability of this medium is very questionable since RCA's announced withdrawal from the player market earlier this year (See LSN May 1984). On the other hand, despite its comparatively low penetration of the consumer market, the future of the reflective optical videodisc looks quite good. In a report published in the August 1984 issue of Videodisc Monitor, a number of sources that track current and projected sales of videodisc players in the U.S. were examined. According to one source, the current estimated population of consumer videodisc players in the U.S. is 630,000- 500,000 CED players and 130,000 reflective optical videodisc players.

The reason for optimism for the future of the reflective videodisc lies in its ability to capture the industrial market currently estimated at 76,000 units, divided among arcade games (30,000), educational use (6,000), and general use (40,000). The reflective optical videodisc monopolizes this market on account of its interactive capabilities and its utilization of the full range of videodisc capabilities including freeze-frame and slow motion. Integration of these features into programs for educational and industrial use allows for some very sophisticated applications.

Among the U.S. organizations that have made large purchases of players for such applications are: American Motors (1,200), AT&T/AIS (350), Digital Equipment Corp. (4,000), Disney Epcot Center (272), Ford-U.S. and Canada (4,200), General Motors (10,500), IBM-worldwide (4,400), Miles Pharmaceuticals (250), and the U.S. Army (3,100).

The 1984 market size for interactive videodisc is estimated to be $66 million, with 54 percent of the applications relating to training, and 23 percent each in advertising/sales/promotion,' public relations and coin-operated applications such as video arcade games. It is expected that by 1993, the market will be over $7 billion.

Another study projects that, in l989 the point-of-purchase and information kiosk market will surpass the consumer --market and become the largest user of videodisc. That report also predicts that the total annual U.S. production of discs will grow from the current rate of 3 million to 24 million by 1990, at which time, roughly 15 million will be for consumers and 9 million for industrial use.

If current plans adhere to schedule, libraries will soon be added to the "industrial market" applications. Library Systems and Services Inc. (LSSI) intends to begin deliveries of its videodisc-based version of the MiniMARC cataloging support system this month. This, and similar systems being investigated by companies such as Carrollton Press, Wilson, CLSI and Geac, use the reflective optical disc as a low cost, high density machine-readable data base publishing medium.

The digital data to be mastered on the disc is recorded on videotape. Proprietary processes, first brought to the market by LaserData of Cambridge, MA, are applied in recording the data on videotape, and processing it to detect errors. Production of the disks from the tape follows regular videodisc production procedures and uses commercial mastering and replication facilities such as those of Sony and 3M. The disks are played on regular laser optical videodisc players with specially developed computer interfaces which enable the disc player to be accessed like any regular computer storage peripheral. On playback, a controller reconverts the analog signal to digital form, and sends the digital data to the computer.

A number of other companies have developed proprietary processes similar to those used by LaserData to encode digital information on analog videodiscs, Firms active in the field include Online Computer Systems, Inc. of Germantown, MD, Reference Technology Inc. of Boulder, CO, and Tomax Inc. of Montreal. These methods which store from 25 to 100 megabytes of digital data per minute of video run time, all require the performance characteristics of reflective optical videodisc Some companies which have investigated the use of videodisc for publishing digital data appear to be holding back. There are at least two reasons for this reticence: 1) they want to be able to assess the efforts of pioneers such as LSSI, or 2) they are concerned lest this use of videodisc be eclipsed by the use of compact audio disks for machine-readable data publishing. Although computing journals have long been predicting the imminent appearance of microcomputer software and data banks in this format (referred to as CD ROM which stands for Compact Disk Read Only Memory) no operational applications have yet been announced. At present, market penetration of the compact audio disk players has not been sufficient to reduce the price of their components to make them attractive as computer storage peripherals, and there are no U.S. production facilities devoted to this media. If experience with the storage of digital data on videodisc is any guide (and it need not be as the compact disc can record such data in digital form rather than requiring its conversion to an analog television-compatible image) it could be two or more years before CD-ROMs are widely available.

Irrespective of the outcome of any battle between the reflective optical videodisc and CD ROM as digital data publishing media, the high degree of activity in other interactive applications of the videodisc technology combined with the consumer market for entertainment discs, should ensure the continuing viability of this videodisc medium.

It is not improbable that a related type of videodisc-the transmissive optical disc-will gain some ground in the U.S. market in the future. Although promised for widespread release in the U.S. for some time, the potential impact of this third videodisc format is yet to be assessed. This format, which is promoted by Thomson-CSF, is currently being supported by the French Government. However, given the size of the established base of reflective optical videodisc technology in the U.S., it does not appear likely that there would be a wholesale defection to the Thomson disc if it does become available here.

Pilot telefacsimile projects in libraries

Several brief (two, four, or six months) pilot telefacsimile projects have been undertaken in the past year, primarily by academic and special libraries. Costs per page were generally higher than the $.30-.50 per page incurred in business applications. That is to be expected because the density of information on typeset pages is more than double that of the average business letter. What is surprising is the wide range of costs reported. A group of special libraries using equipment shared with other departments of the organization achieved a cost of $.54 per page. One academic library achieved costs of $.68 per page and another $.73 per page, but similar academic libraries incurred costs as high as $1.40 per page. A consortium of different types of libraries reported a staggering cost of $7.12 per page.

The special libraries reporting the lowest costs averaged over 1,500 pages per month per machine. The academic library reporting the lowest cost averaged 1,297 pages per month on its machine and the consortium reporting $7.12 per page averaged only 56 pages per month per machine.

There was another difference among the 12 institutions for which costs became available: the special libraries achieving the lowest cost made all transmissions between 11:00 p.m. and 8:00 a.m.; the academic library reporting the lowest cost used a discount telephone service; and those incurring the highest costs made a number of transmissions over regular telephone lines during the daytime. A graphic illustration of the importance of controlling telecommunications costs is that a library transmitting 50 percent more pages than another incurred a telecommunications bill substantially less than half that of the library with the lower volume-a difference which could only partly be explained by the differences in destinations of the materials sent. Telecommunications costs as a percentage of the total cost ranged from under 25 percent to over 70 percent.

Labor costs also varied dramatically, reflecting both care in record keeping and reluctance on the part of some to leave equipment unattended. If the institutions reporting virtually no labor costs are eliminated, the remainder incurred costs as low as $.03 per copy to as high as $.27 per copy. Is it necessary to attend the equipment? Many librarians think so. In the case of some institutions the number of pages requiring retransmission exceeded 10 percent, but on average, the percentage was closer to $ and some institutions experienced much lower rates.

There is not yet sufficient experience with a well managed, fully operational system to determine whether libraries can bring their costs down to a level near $.50 per page. It is clear that costs can quickly get out of hand if the use of the technology is not carefully controlled.

DataPhase scores

DataPhase Corporation scored big twice in the past month. The Chicago Public Library officially accepted the ALIS III hardware and software, paving the way for a major payment and removing the cloud which had hung over the vendor, Only the telecommunications network has not yet been accepted. DataPhase was also awarded the coveted TriLi contract to supply an automated circulation control system to be shared by the Brooklyn, New York and Queens Borough Public Libraries. Together the libraries comprise the largest group of branch libraries in the nation, with over 200 agencies containing 10 million items and circulating in excess of 25 million items a year.

---

Maintained by Marshall Breeding

Copyright 1994-2024