ACRL News Issue (B) of College & Research Libraries 528 / C &R L N ew s Issu es in r e tr o sp e c tiv e c o n v e r s io n By C. Lee Jon es Program C onsultant C ouncil on L ib ra ry Resources Nine recommendations fo r a coordinated program to produce and share machine-readable bibliographic records nationallu. pF rom July 16 th ro u g h 18, a t the Spring Hill Con- ference C en ter in W a y z a ta , M innesota, tw enty- nine individuals focused th e ir atten tio n on a re ­ p o rt, Issues in R etrospective C onversion, p re p a re d by J u tta R eed-Seott, D o ro th y G regor, an d C harles P a y n e .1 T he underlying question d u rin g th e con­ ference w as w h eth er or not th e cause of scholarship could be advanced by a carefully a rtic u la te d p ro ­ gram aim ed at a co o rd in ated ap p ro ach to re tro ­ spective conversion of m a n u al bibliographic rec­ ords into m ach in e-read ab le form by th e research libraries of th e country. S upport of scholarship an d research is th e fu n d a ­ m en tal objective of any retrospective conversion p ro g ram . A req u irem en t for p roviding th a t su p ­ p o rt is an openly accessible, consistent, logical n a ­ tional d atab ase of b ib liographic records reflecting th e n ation's lib ra ry resources. T h ro u g h o u t th e con­ ference it w as clear th a t th e re are no n a tio n a l boundaries to scholarship an d th a t in th e shorter te rm , th e reco m m en d atio n s of th e conference an d subsequent actions tak en should include all N orth A m erican interests, w ith a longer term goal of links to any b ib lio g rap h ic d atab ase in th e w orld. T h e conference discussions ran g ed broadly and the d eb ate assured th a t a w ide set of approaches and concerns w ere aired an d becam e in some w ay a 1This re p o rt is available for $3 (p rep aid only) from th e C ouncil on L ib ra ry Resources, 1785 M as­ sachusetts Avenue, N .W ., W ashington, D C 20036. t l t t t m b p s t w f e a r t of th e recom m endations w hich follow. W hile h e re m a y be altern ativ e approaches to th e p ro b ­ em , th e follow ing represents th e sense of th e group g ath ered a t Spring Hill. G iven th e qu ality and ex­ e n t of in fo rm atio n available (in some cases m ore h a n a d e q u a te an d in others fru stratin g ly sketchy) hese re c o m m e n d a tio n s a re th e strongest sta te ­ ents th a t can be issued a t this tim e. They suggest action on th e p a r t of th e Association of Research L i­ raries as an o rg an izatio n an d , less directly, on the a r t of every research lib ra ry in th e country. Recommendation 1 A c o o rd in a ted retro sp ective conversion (R E ­ C O N ) program is a viable alternative a t this tim e. W h a te v e r p ro g ra m results from these re c o m ­ m endations should no t be view ed as a rep lacem en t for existing local R E C O N form ats. T h e fu n d a m e n ­ ta l objective of such a p ro g ram m ust be th e conver­ sion of m a n u a l records to m a ch in e-read ab le form an d , secondarily, th e u p g rad in g of ex tan t m ach in e records th a t are less th a n full records. This first reco m m en d atio n flow ed from th e dis­ cussion of w h e th e r or not it m a d e sense to m o u n t uch an organized effort a t this tim e. Some argued h a t th e re w as so m uch R E C O N going on th a t it o u ld all be done in th e next few years, despite the a c t th a t th e re is no p la n in place now an d a p p a r­ ntly m u ch d u p licatio n of effort. T h e specific de­ gree of d u p licatio n w as n o t know n, b u t suspected Novem ber 1984 / 529 to be high. Since there are few if any RECON p ro ­ grams searching all shared cataloging service d a ta ­ bases and none sharing records w ith all of them (OCLC, RLIN, W LN and LC), the argum ent was m ade th a t duplication was bound to be high. A m ajor question was w h at was m eant by RE­ CON. In the context of this conference, RECON was lim ited to the original generation of m achine records from older m a n u a l records. T hus, th e nearly clerical tasks of identifying records in a database and attaching a holding symbol w ere de­ lib e ra te ly excluded from any p la n th a t m ig h t result. However, the process of identifying and u p ­ grading to full record status any m inim al record was considered to be an im portant contribution to the quality of resulting databases and so an im por­ ta n t p a rt of a national RECON plan. Recommendation 2 North American research libraries are the focus o f these recommendations and any program that m ay flo w fro m them. However, this focus m ust also include the special resources o f other institu­ tions. A principal reason for this focus is the present state of RECON in North America. M any smaller academic and public libraries have finished, em ­ barked upon, or have near-term plans for the com­ plete RECON of their collections. It was indicated th a t most collections of less th an 250,000 volumes can be converted at reasonable institutional costs using the very large databases of one of the utilities or the services of a commercial vendor. A very high percentage of th e RECO N w ork will consist of m atching records from the database and so require very little original RECON work. However, as col­ lection size grows the costs of RECON become so large th a t it is difficult to devote sufficient institu­ tional resources to accomplish a total RECON. W hile there are few large research libraries th a t do not have some RECON activity either in process or scheduled, it is rare to find one intending to do all extant records because of the num ber of items re q u irin g o rig in al R EC O N w ork an d th e very much higher costs associated w ith this process. In order to assist research libraries w ith their RECON loads, a coordinated program has a very good chance of distributing the am ount of original RE- CON work th a t any one institution would have to do. T h e re is no d o u b t th a t m a n y sm aller n o n ­ research libraries m ust do original REC O N for some p a rt of their collections. It is also probably tru e th a t there is not m uch local pressure on them to convert their special collections unless there is an institutional com m itm ent to convert all records. For the most p art, these libraries will have com par­ atively few unique records to add to a national database. Consequently, in order to expand the n a ­ tional database of RECON records for the benefit of scholars everywhere, it makes sense to focus the national coordinated RECON program on the re­ search library community. Recommendation 3 The Association o f Research Libraries, probably through its B ibliographic C ontrol C o m m itte e , should assume program definition and manage­ m ent oversight responsibilities. There are several models for operation of such a A RL should assume management oversight responsibilities. program w ith in ARL including th e m icroform clearinghouse and certain OMS operations. ARL should also invite the participation of a representa­ tive of IRLA (Independent Research L ibrary Asso­ ciation) during the program definition stages of the process. E arly attention needs to be paid to the ex­ act scope of the RECON problem . Since most of the RECON problem th a t remains to be solved, at least in terms of original RE CON, is located w ithin research libraries, it is reasonable to suggest th a t ARL should take the organizational lead in defining w h at should be done and the stra t­ egy th a t should be used. These recom mendations will be forw arded to ARL as soon as possible for their action. Not all research libraries are members of ARL; several are members of IRLA. To make certain th a t non-ARL m em ber research libraries are p a rt of the program and are urged to contribute to the end result, they should be invited to participate in the Bibliographic Control Com m ittee. There will be other collections th a t should be a p a rt of the p ro ­ gram th a t are not represented by these two organi­ zations, an d th e ir interests should also be ac­ counted for in the definition of the program . W hile th e B ibliographic C ontrol C om m ittee cannot be expected to m anage the program , there are models w ithin ARL th a t m ight be used to p a t­ tern the RECON program . The microform clear­ inghouse effort and certain OMS operations should be examined in order to determ ine the best w ay to handle the daily m anagem ent requirem ents of a program designed to coordinate RECON activities w ithin the research library community. The precise scope of the original RE CON p ro b ­ lem is unknown. A modest and rapidly m ounted ef­ fort should be m ade to determ ine the size of the problem and some indication of how it should be approached. Are there concentrations of records th a t need to be converted? Is the problem tra c ta ­ ble? 530 / C&RL News Recommendation 4 A coordination program fo r R E C O N m ust capi­ talize fu lly upon other R E C O N record-producing activities. There are several projects th a t are already u n ­ derw ay th a t are creating w h at am ount to RECO N records as by-products of their activities. These in ­ clude the N EH -funded new spaper project, th e m a­ jor microform set project, R L G ’s RECON project, and certain preservation projects. Each of these projects produces bibliographic records w hich ei­ ther replace existing m anual records or upgrade in­ com plete m achine form records. It is possible to identify the groups of m aterials th a t are being dealt w ith in these projects and any RE CON program must do so in order to avoid or minimize duplicate record production. By recognizing the contribu­ tions expected from these ongoing activities, the ARL plan for coordinated RECON work will in­ clude these projects and so expand th e productivity th a t can be expected from it. Recommendation 5 The approach recom m ended is to segregate the work by subject based upon the E C classification scheme, or, under certain conditions, based upon certain very strong special collections. Two program s were identified as being useful in identifying institutional strengths based upon sub­ jects as defined by the LC classification scheme: the N ational Collection Inventory Project (NCIP) and th e N a tio n a l Shelflist C o u n t (NSC). Since th e form er is still in its early stages of operation, it is more likely th a t the NSC will prove to be m ore use­ ful in th e short ru n despite certain lim itations. W hile LC inform ation is p a rt of the NSC d ata, only 25 other libraries are included in this 1977 com pilation. NSC can yield initial inform ation on size of research collection, w hile NCIP will eventu­ ally yield collection quality data. Recommendation 6 I t is im portant that any institution choosing to participate in the coordinated R E C O N program m ust agree to produce and share records according to a set o f agreed upon standards. Those standards are based upon the prem ise th a t the fullest possible record properly encoded is to be preferred. Specific standards th a t should be fol­ lowed include: 1. MARC fo rm a t for d a ta encoding an d ex­ change. 2. The N ational Level Bibliographic R ecord- M inimal Level Record standard is the least accept­ able record. 3. AACR2 is preferred for access points. 4. S ubject headings should be LC SH /M eSH com patible. There is no point in p u ttin g together a program for coordinated RE CON unless the resulting rec­ ords can be shared. Sharing records requires an agreed upon set of sta n d a rd s an d d istrib u tio n am ong the database of the large shared cataloging services and other suppliers of bibliographic rec­ ords. In the end, these databases are th e com po­ nent parts of our objective, a logical, consistent n a ­ tional bibliographic database openly available to all citizens. The use of MARC for the exchange of d a ta is as­ sumed. The specification of the M inim al Level Record as the least acceptable standard is intended to spec­ ify the absolute m inim um and not to specify the target against w hich to measure quality. In fact, if support is provided in the context of the program for the production of RE CON original records, lit­ tle or no support should be provided for the p ro ­ duction of m inim al records. Any m inim al record th a t is selected by another institution may force th a t institution to do additional work if it chooses to use only full records in its database. The object of the program is to do as m any original RECON rec­ ords as possible and do them fully once so th a t o th ­ ers m ay sh are th em w ith o u t u n d u e a d d itio n a l work. There is also a need to keep the costs of RE CON under control. Specifying AACR2 as preferred for access points is a case in point. Should all access points and descriptions be required to be consistent w ith AACR2, few if any institutions could afford the tim e required to bring old records up to the new standard. This would am ount to recataloging and not just converting from a m anual to a m achine record. It w ould be possible to convert headings by running them against an AACR2 authority file and dealing only w ith those th a t do not match. W here subject headings are used in a RECON record they should be consistent w ith the two larg­ est controlled subject heading lists, the Library o f Congress Subject Headings and the National L i­ b rary of Medicine’s M edical Subject Headings. It was agreed th a t subjects will be im portant even for the older records, and th a t an effort to assure con­ sistency w ith th e two prim e subject heading stan­ dards was required for the benefit of users. Recommendation 7 The telecom m unication protocols resulting fro m the L inked Systems Project should provide the ena­ bling m echanism fo r the sharing o f records pro­ duced on the several shared cataloging services’ sys­ tems. The linking protocols th a t are now in the testing phase and approaching the im plem entation stages are viewed as th e a p p ro p ria te m echanism s for m aking resulting and other records available to the library user com m unity. It is realized th a t this will take some tim e to im plem ent, b u t it should remain the objective of the library com m unity. Recommendation 8 W hen the Linked Systems Project is fully opera­ N ovem ber 1984 / 531 tional, access to records will be diam etrically en­ hanced. H ow ever, access to all original or u p ­ g ra d e d records re su ltin g fr o m a c o o rd in a te d R E C O N program should be provided through an L C distribution m echanism . Any LC records in p u t b y a n o th e r o rg a n iz a tio n a c c o rd in g to c e rta in guidelines specified by LC could be accepted as LC MARC records, and so, distributed as p a rt of the MARC D istribution Service, and m aintained by th e L ib rary of Congress. T he L ib rary of Congress agreed to investigate the possibility of an LC distri­ bution m echanism . P articipants recognized th a t th e w idest possible distribution w ould come from using an LC distri­ bution m echanism . All subscribers w ould receive the records and there are no limits, other th a n fi­ nancial, to those who m ay subscribe. T he com m er­ cial sector will thus be served as readily as th e not- for-profit sector. It w ould be particularly desirable to have th e LC cataloging records included in the MARC D istribution Service w here they w ould be p a rt of a continuing m aintenance program . Recommendation 9 T he A R L R ib lio g ra p h ic C o n tro l C o m m itte e should explore a variety o f fu n d in g operations fo r the support o f a coordinated R E C O N program. Individual projects should be packaged for fu n d ­ raising purposes. In addition, a strategy of assess­ m ents of research libraries should be explored. F i­ nally, ap p ro p riate staff from Title II-C and N EH should be advised th a t a coordinated RE CON p ro ­ gram is being prepared. RE CON activities have received sporadic sup­ p o rt for several years. Much has been done w ith lo­ cal funding in efforts to im plem ent online circula­ tio n sy stem s t h a t r e q u ir e d m o re o r less fu ll bibliographic records as raw m aterial, and there is bound to be m ore local support of RE CON in the years to come. A national coordinated strategy for retrospective conversion, w hether fully or partially funded or not, w ould provide th e context in w hich in stitutions could ap p ro ach th e ir ow n RE CO N projects, know ing th a t they w ould be m aking up a contribution to th e national REC O N effort. Thus, a w orst case of no extram ural support for RE CON still calls for a logical, coordinated RECO N plan. There are, how ever, m any foundations w hich m ay be interested in specific pieces of th e RE CON problem . It w ould be useful, for example, to p u t together a package th a t m ight be interesting to the G etty F oundation in the area of art and architec­ ture. O th er foundations m ay be interested in other pieces of th e knowledge spectrum . It w ould not be terribly difficult, once d a ta was exam ined from the NSIP, to p u t together some subject assignment sug­ gestions for a lim ited num ber of institutions, to se­ cure th eir com m itm ent to the concept, and to seek support for a special package of m aterials. The n a ­ tu re of th e support should be such th a t each p artici­ p atin g institution is investing in the project rath e r substantially. Foundations are m ore likely to be in­ terested in providing m atching funds th a n they are in supporting all the institution’s costs relative to R ECO N . Support should be sought w ith in these lim itations. In a m ore broadly based program for generating support for R EC O N , ARL should consider a p ro ­ gram of assessments of research libraries in order to accum ulate resources th a t could be used as m atch- Objective: A national database openly available to all citizens. ing funds for the RE CON effort. Some institutions w ill be able to m ake cash paym ents to a project- specific fund. Others w ill not be able to do m uch m ore th a n to allocate a specific sum w ith in their o p eratin g budgets as m a tch in g su p p o rt for R E ­ CON. These non-cash com m itm ents are likely to be in th e form of staff an d o th e r resources and should be viewed as an acceptable alternative to cash com m itm ents. Again, such a resource pool m ay a ttra c t m atching attention from the founda­ tion and federal funding com munities. Both Title II-C and N EH have funded RECO N projects in a less th a n coordinated w ay over the past several years. Both agencies should be alerted to th e fact th a t there is now an effort to produce a logical co o rd in ated p la n for RE CO N activities. A lerting these tw o program s to w h a t is com ing should allow them to capitalize upon th e plan in th eir support of RE CON proposals. T he foregoing constitute the central recom m en­ dations of the three-day conference. As one m ight expect, there w ere m any other recom m endations th a t did not receive such w ide support or w hich h ad poor specific fits in th e program recom m ended above. T he most useful of these recom m endations are su m m a riz e d as m iscellanous re c o m m e n d a ­ tions. Miscellanous recommendations O p tic a l C h a ra c te r R ecognition (OCR) te c h ­ niques should be explored in light of some new de­ velopm ents in th e field. Given th a t most of RE- CO N is th e capture of d a ta th a t already exists in a variety of p rin t form ats and th a t th ere have been some interesting recent developments in OCR and th e controlling softw are, a renew ed exam ination of O CR technology for purposes of supporting R E ­ CON should be undertaken. T he Council on L i­ b rary Resources has com m itted itself to such an ap ­ praisal. One of the products of th e RECO N program will 532 / C& RL N ew s be the conversion of records th a t w ere originally the p ro d u ct of LC cataloging. It was suggested th a t all of these records should be flagged and sent to LC by th e shared cataloging services as a service to L C . LC representatives agreed to consider the useful­ ness to LC of this suggestion. W hile th e Linked Systems Project will result in operating links betw een the L ib rary of Congress, th e Research Libraries G roup, and th e W ashing­ ton L ib rary N etw ork, it will be some tim e before O C L C can become a p a rt of th e technical link. M i­ cro en h an cer or sim ilar techniques using m icro ­ com puters should be developed for searching sev­ eral databases in th e R EC O N process. Since one of th e objectives is to red u ce d u p licativ e effort, it makes no sense to search only one database w hen there is some likelihood th a t sim ilar w ork m ay al­ ready have been done on one or m ore others. This suggestion m ay require m ore softw are w ork on the p a rt of th e targ et databases th a n they are w illing to do, b u t th ere was encouragem ent to explore this avenue as a short-term solution to the lack of o p era­ tional links am ong th e utilities. These recom m endations form th e essence of a nationally coordinated program for retrospective conversion of p rin t form bibliographic records. It is a program th a t has the chance of reducing the ag­ gregate costs of th e RE CON process and securing funding for m aking a very large dent in the inven­ to r y of re c o rd s t h a t n e e d to b e c o n v e rte d to m achine-readable form in order to b etter support the w ork of th e scholarly com m unity. ■ ■ E d ito r’s note: No copyright is claim ed on this arti­ cle, w hich the author w rote as part o f his official duties as an em ployee o f the U.S. governm ent. M a n a g in g a u to m a tio n fo r results: C om p letin g task s w h ile p a rticip a tiv e p la n n in g p r o c ee d s B y E llen G. M iller Director, L ibrary Systems D evelopm ent University o f C incinnati Planning for automation w ith the end-user in mind. F or m any university and research libraries, the lapsed tim e betw een starting library autom ation p la n n in g a n d se c u rin g fin a l to p m a n a g e m e n t b u d g e t signatures is several m onths. Perhaps it even takes years. The com petition in higher educa­ tion for scarce resources means th a t lib rary au to ­ m ation m anagers and C EO librarians m ust create a careful case for lib rary autom ation. O ne m ethod for creating th a t case, tim e-consum ing b u t politi­ cally and psychologically helpful, is p articipative planning. As used in this article, participative planning goes beyond the recent lib rary literatu re concern­ ing participative m an ag em en t1 because it specifi- 1For a useful sum m ary, see Nicholas C. Burckel, “ P a r tic ip a to r y M a n a g e m e n t in A cadem ic L i­ b raries: A R e v ie w ,’’ C ollege a n d Research L i ­ braries 45(January 1984):25-34.