ACRL News Issue (B) of College & Research Libraries 650 / C &R L N ew s for the nation as a whole if vital historical informa­ tion as well as electronic information, which is just in the development stages, should be preserved. L i ­ brarians must continue to sponsor programs to in­ crease the public’s awareness of the urgency of the preservation problems. Public and private funding sources should be mobilized to support adequate preservation programs to conserve the nation’s in­ fo rm a tio n b a s e . — H a n n e lo r e B. R a d e r , A C R L P resident. The expanding universe of special formats B y M a rg a re t E . G allow ay A ssociate D irector o j L ib r a r ie s N orth Texas State University K e n n e th L a v e n d e r University B ib lio g rap h er N orth Texas State University G eo rg e M itchell D irector, M edia L ib ra ry N orth Texas State University a n d W illiam F lo y d Systems D ev elo p m en t L ib ra ria n N orth Texas State University Some local strategies fo r collection developm ent of co m p uter software. I n the fall of 1982, the North Texas State Univer sity Libraries received several requests for com puter software. Since the libraries had never pur chased this type of material, the administratio appointed a com m ittee to investigate the matte and make recommendations. Two issues were o immediate concern: the considerable expense in volved in the purchase of such materials includin the necessity for special handling, and the need t provide accessibility of the material to our patrons. Both of these concerns had far-reaching implica tions. It was obvious that such materials tended t be extremely expensive, difficult to evaluate fo p u rch ase, tim e-co n su m in g to ca ta lo g , and r e quired special, diverse equipment to access. For in sight into the problems, the committee decided t ­ ­ ­ n r f ­ g o ­ o r ­ ­ o review the literature then available, but the review provided only limited assistance. Next the com m it­ tee issued a questionnaire to be sent to a select group of libraries in the United States and C a n ­ a d a .1 Analysis of the responses to the questionnaire simply accented the general lack of information on how to deal with machine-readable formats as li­ brary m aterials.2 1The questionnaire was sent to 100 academic and research libraries: the 50 libraries identified from statistics of Southern college and university li­ braries published by Louisiana State University L i ­ brary; and the 50 largest libraries from the current A R L listing not included in the first 50. Forty-eight libraries provided usable responses to all questions. 2 P E C Kit #123, M icro co m p u ter S o ftw a re Poll- N ovem ber 1986 / 651 At this point, the committee agreed that it must take the philosophical stance that information, re­ gardless of format, should be collected by an aca­ demic library if it is important to the educational goals of the university. Since this decision would impact the outdated collection policy of the NTSU Libraries, it was proposed that the collection policy be revised to reflect the need to collect computer software. As it worked on its recommendations, the committee determined that no previous at­ tempt had been made to define the primary clien­ tele of the Libraries. As a state institution, the L i­ braries receive all funding as an appropriation from the state legislature, based on a formula of student credit hours that allocates more money for doctoral and master’s hours than for undergradu­ ate hours. No funds are directly allocated for fac­ ulty, staff, or the community whose taxes support the institution. It is also true, however, that without the re­ search of the graduate faculty there would be no master's or doctoral students and that the commu­ nity has always had access to the Libraries’ collec­ tions. The committee defined primary clientele as “Persons with an official connection t o . .. (the) Uni­ versity, including students, faculty, and staff.” The definition of primary clientele was immediately adopted as a part of a policy that defined the prin­ ciples for services in the Libraries and thus also adopted as part of the revised collection develop­ ment policy. While seemingly a simple decision, it formed the basis for all recommendations made by the committee. For example, the committee deter­ mined that having material no one could access, or having material simply to have the format, was not appropriate to these principles. More specifically, it opposed the idea of purchasing computer soft­ ware that could be used by only one faculty mem­ ber for a specific research project. Thus these prin­ ciples have already had a p p lic a tio n in the development of the Libraries’ collections. Selection guidelines for computer software and data files now have been written and appended to the general collection development policy of the Libraries.3 The guidelines basically state that be­ cause of the high cost of these materials, and the importance of developing a well-used collection, all decisions to purchase are made with great care. Either a preview of the material is performed by the faculty requesting it, or reviews are located cies in A R L L ibraries (April 1988), was not avail­ able at the time of the survey. The results of its own survey, however, indicated that only 34 % of the responding libraries had policies or were in the pro­ cess of creating them. “Moreover, there is some concern that by not assimilating this new medium, libraries might be abdicating a significant role in the information age.” 3Appendix C and D of the North Texas State Uni­ versity Libraries’ Collection Development Policy are appended. that adequately assess its potential. The cost of cat­ aloging and storage is taken into consideration. Materials for microcomputer use and for both mini- or main-frame computer use must be shown to support curriculum or research needs. F r e ­ quently these requests relate directly to the Univer­ sity’s stated goals and mission to respond to the growing hi-tech needs of the Dallas/Fort Worth metroplex. Funding for purchases has come from a variety of sources: the computing center, gifts, and the li­ brary materials fund. It is now clear from our expe­ rience that each party to the purchase must work closely and cooperatively in order to insure the best use of limited funds. To track the location of soft­ ware on campus, and yet to make the tools of in­ struction and research as widely available as possi­ ble, the Libraries are working with the University Computing Council toward a system that will ex­ ert control over the proliferation of software and databases on campus. A master backup copy of mi­ crocomputer software and data files is housed in the Media Library, but the circulating copies and documentation are made available in the area where the material is to be used. Software and data files for use on the mini- and main-frame computers in the computing center are housed in the center’s tape handling facility. The facility has its own management and utilization control system to locate tapes and keep data on them from deteriorating, but this system does not replace cataloging by the Libraries. All library- cataloged materials in the computing center are listed in the Virginia Tech Library System (VTLS) online catalog of the Libraries and are accessible to the Libraries’ primary clientele through the com­ puting center. Indeed, anyone with a terminal anywhere on campus can dial into the center through the cam ­ pus local area network and use any of the software or data files cataloged in the Libraries. Copies of documentation are also cataloged and available in the appropriate reference or reserve area of the L i­ braries. Standardization, user convenience, and copy­ rights and license agreements were concerns that required a university-wide approach. Because the Libraries have defined the primary clientele, es­ tablished a flexible collection development policy, and laid the groundwork for cooperative acquisi­ tion and use of software and data files, they are now part of the university’s decision-making pro­ cess. Thus, the Libraries have created means which respond to the access needs of patrons, using university-wide facilities, rather than attempting to rely mainly upon their own. While these means may not be traditional library methods of opera­ tion, they have proven effective. By being con­ cerned at a relatively early date about the issues surrounding computer software and data files, the Libraries have a strong position in building a coop­ 652 / C &R L News erative campus-wide software and data file net­ work. A PPEN D IX C NTSU Guidelines for selection of computer software As the technology of data storage changes, research li­ braries must be committed to providing access to all types of information regardless of format. Usually, one con­ siders computer software as the programs used to operate a computer and manipulate stored, coded data. For the purpose of these guidelines, the committee included not only the former meaning but also the concept of machine readable data files (hereafter M R D F ). M R D F is any col­ lection of data, text and/or numeric, that can be accessed and/or manipulated by computer. For these guidelines, the L ibraries' primary clientele remains students, faculty, and staff. A. Factors affecting acquisition Before acquiring computer software, one must consider several issues. These issues include consortium and con­ tractual agreements and the need to provide access to this type of informational source. Standard collection devel­ opment considerations remain important. However, there are additional factors specific to this format of data stor­ age which affect decisions concerning acquisition. 1. S u bject scop e. As with other materials, textual and numeric data files must be evaluated in relationship to the instructional and research interests of the academic com ­ munity. 2. User n eed . T h e information needs of academic users must be examined to determine if primary data is needed in this format. Some users may need to manipulate pri­ mary data, rather than use already manipulated data. Conversely, some users may not need to deal with data manipulation and its attendant problems. 3. U niqueness o f d a ta . As information sources prolifer­ ate, it becomes imperative to examine the uniqueness of data in this format. If the data is also available in other formats, in which format(s) should this data be acquired? There often are reasons to have access in more than one format. 4. A u thority o f d a ta . T h e origins of the data should be examined to assess authoritativeness and completeness. D ata representing research in progress which might even­ tually undergo significant change is perhaps of limited value. 5. C o n fid en tia lity o f d ata. Some textual and numeric data files may include proprietary or restricted inform a­ tion and consequently may be limited in terms of the L i ­ braries’ primary clientele. Some files may have suppressed information to insure confidentiality. 6. Physical fo r m a t . In these times of rapid technological change, a variety of physical formats may be available. Some materials available on magnetic tape are also avail­ able in floppy disc, disc pack, or optical disk format. 7. O nline access. Some sources are available only on­ line. For business and industry statistics, online access pro­ vides immediate forecasting capabilities. W hile most data sources are not available in this fashion, it is conceivable that more will become available as the need and demand for this type of information grows. 8. D o c u m en ta tio n , consulting an d in stru ction al su p­ p ort, a n d so ftw are. Using textual and numeric data files requires adequate docum entation so users can extract needed information. Complex files may require staff to as­ sist users with file manipulation. Hardware and software issues must also be addressed, including availability of computing facilities and software within the campus com ­ munity. 9. E xten t o f u tiliz a tio n . T h e extent to which the m ate­ rial is expected to be used as judged by expressed interest, number of classes to which appropriate, quality of con­ tent, and technical quality. 10. D u p lication . The need for duplication of heavily used items and for replacement of damaged items. 11. C o o p e r a tiv e acqu isition . Purchase decisions should consider the accessibility of materials within the North Texas area. 12. C on tracts. All contracts should be negotiated with provisions for the broadest possible use by the Libraries’ primary clientele. B. Documentation T h e successful use of computer software as well as suc­ cessful searches of M R D F often depend upon understand­ ing the methods by which the software works and the methods by which the data was coded. This implies that the documentation must be available to the users. T h e re­ fore, the Libraries should serve as the central repository of information related to any computer software acquired. These items would include but are not limited to: 1. published directories of data files and guides to the holdings of data archives which identify the existence and availabilitv of these resources; 2. core collections of periodicals and monographs de­ voted to the utilization of these materials; and 3. code books, systems manuals, and other documenta­ tion, if appropriate. This documentation should circulate to all users. H ow­ ever, because of heavy demand, it may be necessary to limit the access or length of the borrowing period for some items. C. Policies 1. Generally, the NTSU Libraries will acquire com ­ puter software only wh ere the magnitude of the research project or teaching warrants the expenditure of time, ef­ fort, and money. 2. T h e Libraries will also maintain annually updated M R D F on a limited basis. 3. Requests for items not broadly or intensively useful will receive a lower priority. 4. T h e NTSU Libraries will not acquire M R D F that, under contractual agreements, can only be used by one person. Any M R D F obtained must be, at the very least, in some way accessible to any member of the Libraries' pri­ mary clientele. 5. Also excluded are such M R D F that by virtue of their content or format are not likely to be of interest to any other user, such as files consisting of special purpose aggre­ gations of data for one user. 6. The Libraries will not purchase or otherwise acquire software for microcomputers that would normally be ex­ pected to be owned by individuals for personal use, unless the software fulfills a research or work routine function. 7. Only games with sufficient educational or instruc­ tional value for use in coursework will be purchased. 8. Final decisions regarding purchase will rest with the Libraries. 9. Access charges, if any, will be the responsibility of the user not the Libraries. T h e Libraries should work for the elimination of access charges involving computer software acquired through Libraries' funds. 654 / C &R L News A PPEN D IX D Handling m icrocom puter software 1. Microcomputer software may be acquired by one of the following methods: a) purchase from Media Library funds; b) transfer from software purchase funds adminis­ tered by the Computing Center in cooperation with the University Computing Council and its Software Subcom ­ mittee; c) other gift. 2. Microcomputer software is added to the collection and officially cataloged through the normal routines. 3. T h e first copy of any software purchased, including both documentation and discs, is housed as an archival copy in the Media Library for use only to make authorized copies for circulation or other use. 4. W hen a software package is purchased through the Computing Council for use in a laboratory, sufficient cop­ ies of software and documentation must be purchased for this use and costs will be borne either by the Computing Center or the D epartm ent sponsoring the laboratory as agreed upon by the Computing Council. L aboratory cop­ ies will be marked as Media L ib ra ry copies, and added to the uncataloged inventory. 5. T h e Computing Center routinely makes available to the Media L ib rary all information (brochures, specifica­ tions, etc.) used in purchasing any particular software, in­ cluding also a copy of the D epartm ental Copy of the Pu r­ chase O rd er and a copy of any license signed by the Com puting Center. 6. T h e Computing Center advises with the Media L i ­ brary with regard to licensing prior to purchase. 7. All discs housed in the Media L ibrary are suitably packaged for circulation in order to achieve the highest level of protection for the application: fitted with rein­ forcement rings, write protected, and labeled to direct us­ ers to avoid placing in magnetic fields, extreme heat or cold. 8. W h en copies of a software program purchased or made for laboratory use are no longer needed in the labo­ ratory, they will be returned to the Media L ib rary for housing or withdrawal. Library and information science collections By M a rg a re t E . G allow ay A ssociate D irector o f L ib ra ries N orth Texas State University A survey on financing, housing, and staffing of collections fo r Schools of Library and Information Sciences. I n the spring of 1986, a survey questionnaire was prepared and mailed to the 56 ALA -accredited li­ brary schools in the United States. T h e purpose of this survey was to identify the source of financing for and the location of the collection which served each accredited library school. A survey with this particular emphasis has not been reported in the literature for a number of years. F ifty of the 56 questionnaires ( 8 9 % ) were co m ­ leted and returned; of the returned q u estio n ­ aires, 49 could be evaluated as planned. T h e ques­ ionnaire was comprised of four m ajor questions hich examined the source of funding, housing, nd servicing of the library school collection. R e ­ ponses to the four questions are reported below. ll 49 respondents answered the initial two ques­ p n t w a s A