1983,30(1),35-43 COMPUTERIZATION OF THENATIONAL UNION CATALOGUE OF SCIENTIFIC SERIALS...

9
Annals of Library Science and Documentation 1983,30(1),35-43 COMPUTERIZATION OF THE NATIONAL UNION CATALOGUE OF SCIENTIFIC SERIALS: PLANNING ; AND ORGANIZATION OF INPUT DESIGN* Describes the development towards compilation of the National Union Catalogue of Scientific Serials. Several constraints in processing the Union Catalogue data by conventional or manual method are mentioned. Objectives of computerization are highlighted. Analyses the feature elements of serial publications keeping in view the objectives. Principles of designing codes of feature elements are elaborated. Principal input designs and their functions are explained. 1. ThlRODUCTION A national union catalogue of scientific serials is an indispensable tool to support the document supply model developed by Insdoc, The fundamental philosophy of this model is to help in development of an effective national interlending system and to exploit the serial resources of the country. Not only this model will support lnsdoc's document supply service, but also help other important institutions to play an active role in the decentralised docu- ment supply system for the country. The compilation of the serial data would also be an essential requirement for a national stock taking of the serial resources in the country and would supply valuable data for planning and funding acquisition programmes. With these objectives in view, the project has been given high priority and has been planned to complete the basic data collection in a reasonable span of time which involved handling of massive holdings data of scientific serials. The compilation and publication of the National Union Catalogue of Scientific Serials * Based on a paper submitted at the Seminar on the Application of Modern Technology to Bibliographic Control Services. New Delhi 10-11 March 1981. Vol 30 No 1 Mar 1983 S.N. DUTTA, R.N. NEOGI, K.L. GOGIA, Y.K. SHARMA, A. SUBBA RAO, B.R. KESHAVA RAO, (MRS.) RAMESH ARORA Insdoc New Delhi - 110067 has therefore been an on-going project of Insdoc, beginning from 1965. The plan envi- saged to bring out the Catalogue has been on schedule and also been well monitored. The work carried out so far has been in two phases. In the first phase, during 1965 to 1978, the main tasks were of the holdings data collection after physical varification for accuracy, editing, organization of several index files and publica- tion of Catalogues/Regional Union Catalogues. The data represented the holdings of libraries with fairly significant build up of serials collec- tion. This work was handled from 1965 in two stages. The first stage without computer aided processing was carried out during 1965 to 1971 and the second stage from 1973 to 1978, computer facility has vbeen utilized for data processing. In the first phase till 1978, seventeen volumes of catalogues containing holdings data of about 30,000 scientific serials in 530 libraries located in different parts of the country have been published. A computer data base has also been created containing information of 24,000 titles with holdings information of 330 libraries. In the second phase since 1978 it has been planned to compile revised and up-to-date hold- ings data from 530 libraries and to collect hold- ings data of another 270 libraries for bringing out the National Union Catalogue of Scientific Serials. 2. PLANNING FOR COMPUTERIZATION It has been a very useful experience in bringing out twelve volumes of Catalogues/ Regional Union Catalogues during 1965 to 1971 by conventional method. The limitations of handling the data by this method are outlined: (1) holdings data collected from various re- gions cannot be processed simultaneously; 35

Transcript of 1983,30(1),35-43 COMPUTERIZATION OF THENATIONAL UNION CATALOGUE OF SCIENTIFIC SERIALS...

  • Annals of Library Science and Documentation1983,30(1),35-43

    COMPUTERIZATION OF THE NATIONAL UNIONCATALOGUE OF SCIENTIFIC SERIALS: PLANNING

    ;

    AND ORGANIZATION OF INPUT DESIGN*

    Describes the development towards compilationof the National Union Catalogue of ScientificSerials. Several constraints in processing theUnion Catalogue data by conventional ormanual method are mentioned. Objectives ofcomputerization are highlighted. Analyses thefeature elements of serial publications keepingin view the objectives. Principles of designingcodes of feature elements are elaborated.Principal input designs and their functions areexplained.1. ThlRODUCTION

    A national union catalogue of scientificserials is an indispensable tool to support thedocument supply model developed by Insdoc,The fundamental philosophy of this model isto help in development of an effective nationalinterlending system and to exploit the serialresources of the country. Not only this modelwill support lnsdoc's document supply service,but also help other important institutions toplay an active role in the decentralised docu-ment supply system for the country. Thecompilation of the serial data would also be anessential requirement for a national stock takingof the serial resources in the country and wouldsupply valuable data for planning and fundingacquisition programmes. With these objectivesin view, the project has been given high priorityand has been planned to complete the basicdata collection in a reasonable span of timewhich involved handling of massive holdingsdata of scientific serials.

    The compilation and publication of theNational Union Catalogue of Scientific Serials

    * Based on a paper submitted at the Seminaron the Application of Modern Technology toBibliographic Control Services. New Delhi 10-11March 1981.

    Vol 30 No 1 Mar 1983

    S.N. DUTTA, R.N. NEOGI, K.L. GOGIA,Y.K. SHARMA, A. SUBBA RAO, B.R.KESHAVA RAO, (MRS.) RAMESH ARORA

    Insdoc New Delhi - 110067has therefore been an on-going project ofInsdoc, beginning from 1965. The plan envi-saged to bring out the Catalogue has been onschedule and also been well monitored. Thework carried out so far has been in two phases.In the first phase, during 1965 to 1978, themain tasks were of the holdings data collectionafter physical varification for accuracy, editing,organization of several index files and publica-tion of Catalogues/Regional Union Catalogues.The data represented the holdings of librarieswith fairly significant build up of serials collec-tion. This work was handled from 1965 in twostages. The first stage without computer aidedprocessing was carried out during 1965 to 1971and the second stage from 1973 to 1978,computer facility has vbeen utilized for dataprocessing.

    In the first phase till 1978, seventeenvolumes of catalogues containing holdings dataof about 30,000 scientific serials in 530 librarieslocated in different parts of the country havebeen published. A computer data base has alsobeen created containing information of 24,000titles with holdings information of 330 libraries.

    In the second phase since 1978 it has beenplanned to compile revised and up-to-date hold-ings data from 530 libraries and to collect hold-ings data of another 270 libraries for bringingout the National Union Catalogue of ScientificSerials.

    2. PLANNING FOR COMPUTERIZATION

    It has been a very useful experience inbringing out twelve volumes of Catalogues/Regional Union Catalogues during 1965 to 1971by conventional method. The limitations ofhandling the data by this method are outlined:

    (1) holdings data collected from various re-gions cannot be processed simultaneously;

    35

  • DUTTA AND OTHERS

    (2) final editing and the preparation of thepress copy of the catalogue can be under-taken in a manner of only one catalogueat a time;

    (3) information collected goes out of datebefore the press copy of the catalogue isfinalised for publication; and

    (4) after the press copy of the catalogue isfinalised, any additional and revised infor-mation received from participating librariesin the form of supplement, cannot be in-corporated immediately unless revisededition of the catalogue is brought out,which is not planned at the moment.

    In addition, it would be a stupendous taskto prepare the press copy of the National UnionCatalogue containing the holdings informationof about 40,000 titles on completion. Moreover,while processing the press copy for srcceedingcatalogues, it has been noticed that about 50 or60 per cent of titles found common in theearlier catalogues. This percentage of commonentries has been on increase with the publicationof subsequent catalogues. The time taken forediting of common titles, to some extent, can besaved but the drudgery and work-load of routineprocedure connected with the preparation ofpress copy could not be avoided.

    In view of the above facts, computerizationof the project has been given serious considera-tion. Following objectives have been kept inview while planning the operation:

    i) to create a data base of scientific serialsheld in participating institutions in machinereadable form, e.g. on magnetic tapes/disks;

    ii) to expedite the publication of the forth-coming regional catalogues by repeated useof common entries of the data base;

    iii) to undertake input preparation of the hold-ings information simultaneously receivedfrom various regions;

    iv) to incorporate required feature elements inthe coded form in the entries so as to facili-tate to generate different files out of thedata base, e.g. classified, sponsor, countryof publication indexes, individual libraryholdings, and holdings of group of special-ised libraries etc.

    v) to undertake revision and updating of thedata continuously on magnetic tapes/disks

    36

    in order to consolidate the information;vi) to obtain the consolidated print-out of the

    National Union Catalogue immediatelyafter completion of input preparation ofdata collected from participating libraries;and

    vii) computer print-out can be used as masterfor photolitho printing.

    Some other aspects, such as, pattern of thecatalogue, output format of the main part andindexes, form of entry etc., have been examinedin detail while developing the input design. Ithas been felt absolutely necessary to identifythe feature elements of the entry which wouldplay an important role in fulfilling the objectivesof computerization.

    3. PATTERN OF THE CATALOGUE A.NDTHE FEATURE ELEMENTS

    The main part of the catalogue consists ofmain entries with holdings and cross-referenceentries, arranged in one alphabetical sequence bytitle of the serials. Filing of the titles follows aset pattern of rules as followed in ''World Listof Scientific Periodicals", 4th ed, 1960. How-ever, acronyms are taken as words and alpha-betised accordingly. International conferencesand symposia are rendered under their officialEnglish names, and arranged alphabetically inthe main part.

    The elements of the main entry are i) titleof the serials; ii) sponsoring body; iii) imprint;iv) bibliographical note; and v) library symbolswith respective holdings information. Theseelements are to appear in the same order in eachmain entry. All these elements in an entry arerequired to be related with the title and to befiled alphabetically according to the title. Forthe purpose of generation of index files, titlesonly have to be selected from the entries. Tobuild up the individual libraries or holdings ofgroup of specialised libraries files, the relevantentries have to be identified by the titles only.Therefore, title is considered to be main or keyfeature element. Similarly the sponsoring body,the subject heading, and the country of publi-cation (country heading) are to be identifiedfrom the entries and are termed as sub-featureelements for the generation of files of classified,sponsor, and country of publication indexes

    Ann Lib Sci Doc

  • COMPUTERIZATION OF THE NATIONAL UNION CATALOGUE OF SCIENTIFIC SERIALS

    respectively. It may be a complicated situationfor machine to identify and to select the titlesfor the generation of different files, becausethese feature elements are variable in lengthand character. Therefore, in any machine read-able system, the feature elements are requiredto be identified by fixed-length characters atthe fixed position in the input unit.

    Principles of designing codes of variousfeature elements of serials publications havebeen explained by Porter (1) in computerizationof the British Union Catalogue of Periodicals:New Periodical Titles (BUCOP:NPT). In addi-tion, some working experience in the datapreparation of BUCOP : NPT, immenselyhelped to adopt almost similar principles indesigning codes of feature elements for theNational Union Catalogue.

    4. DESIGNING OF CODESConsiderable time and effort were involved

    in designing codes of feature elements, e.g.1) Title; 2) Sponsor; 3) Subject heading and4) Country of publication.

    41 TitleCo·de

    Title code or Record code is designed onthe principle that it should be unique fixed-digited code for each title, and the applicationof the code will also help in filing the entriesaccording to a desired alphabetical pattern."ASTM Coden for Periodical Titles" was thoro-ughly scrutinised before developing a newsystem. "ASTM Coden" provides a six-digitedalpha code which is unique for each title, butfails to satisfy the requirement with regard toits role for filing the titles according to desiredalphabetical pattern. Therefore, it is consideredto be worthwhile to design a new code whichcan serve the purpose of a dual role as an iden-tifier and a record sorter. For this purpose"World List of Scientific Periodicals (WLSP)4th ed, 1960", which contains information ofabout 80,000 titles is taken as base. In addition,about 20,000 titles covered in twelve volumesof catalogues published during 1965 to 1971are transferred on cards and consolidated intoa central file (CF) and about 60% titles of CFare found to be common with the WLSP. ASeven-digited alpha-numeric code has been

    Vol 30 No 1 Mar 1983

    designed for title code or record code. Thefirst digit of the code consists of a letter whichrepresents the first significant word of thetitle followed by six-digited numeral. This setof numerals determines the position of thetitle in relation to other titles in the samealphabet. The titles in the CF, already arrangedin the desired pattern are assigned codes inrelation to the titles of the WLSP in an ascend-ing order of numerals. In order to accommodatenew titles, adequate gaps in numbers in betweentwo titles have been provided for future use(Fig.l). Also to overcome the bottleneck toaccommo

  • Fig.1

    Code

    A 103470A 104220A 106470A 107620A 107820

    A 108020A 108320A 108570A 108720A 108820A 109220A 1(J9320A 109420A 109520

    A 109620

    A 109650A 109820

    A 109920

    A 110120

    A 110220

    A 110420

    A 110520A 110530

    A 110540

    A 110640

    A 110920

    DUTTA AND OTHERS

    TITLE CODESample Data

    Title of Serials

    ACTIVITIES AT TURRIALBAACTULIDADES BIOLOGICASACTUALITES BOTANIQUESACTUALITIES PHARMACOLOGIQUESACTUALITIES SCIENTIFIQUES ETINDUSTRIELLESACUSTICAADA FORECASTADANSONIAADDICTIONSADDISONIAADHAESIONADHESIVES AGEADHESIVES AND RESINSADIRONDACK COTTAGE SANATO-RIUM (SARANAC LAKE, N.Y)REPORTS. SEE ANNUAL MEDICALREPORT, TRUDEAU SANATORIUMADMINISTRATION REPORT. SEEREPORTADMINISTRATIVE MANAGEMENTADMINISTRATIVE MEDICINE-TRANS-ACTIONS OF THE CONFERENCEADMINISTRATIVE SCIENCE QUAR-'fERLYADRENAL CORTEX-TRANSACTIONSOF THE CONFERENCEADVANCE ABSTRACTS OF CONTRI-BTJTIONS OF FISHERIES AND AQUA-TIC SCIENCE IN INDIAADVANCE REPORT, BUREAU OFMINES, QUEBECADVANCE THERAPYADVANCED ENERGY CONVERSION.SEE ENERGY CONVERSIONADVANCED MANAGEMENT. SEE SAMADVANCED MANAGEMENT JOURNALADVANCED MANAGEMENT JOURNAL.SEE SAM ADVANCED MANAGEMENTJOURNALADVANCEMENT OF SCIENCE.

    43 Subject Heading CodeUniversal Decimal Classification scheme has

    been used to classify the tides of serials. UDC

    38

    numbers consisting of digits, symbols and punc-tuation marks cannot be sorted in a logicalsequence by computer. Therefore UDC num-bers require conversion into machine readableform by a fixed-field code of uniform charac-ters. One UDC schedule (Abridged EnglishEdition 1961; 3rd ed; B.S. 1000A 1961) hasbeep completely coded by six-digited numericcode numbers in the sequence of main/subclasses. Adequate gaps in numbers in betweenpreceeding and succeeding class numbers havebeen provided in the schedule, so that suitablecodes can be assigned for further division ofthe class numbers. In view of the subjects tobe covered, in the National Union Catalogue,provision for accommodating majority ofnumbers has been kept under the main classes of5 and 6 of UDC.

    A broad outline of the block code numberallotted against block class number of UDC isgiven below:-

    Block class Number Block code Number

    o to 4995 to 535.9,

    536 to 549;955 to 579.958 to 599.9

    6 to 621.039.7621.039.8 to 629.67

    63. to 663.67663.8 to 691.189

    691.2 to 999

    000001-099999100000-199999200000-299999300000-399999400000-499999500000-599999600000-699999700000- 799999800000-899999900000-999999

    In generation of the classified file the titleshaving same subject heading codes are consoli-dated and are sorted alphabetically. The com-mon code is converted in the form of UDCnumber with appropriate subject heading withthe help of a subject heading dictionary file(Fig.3 ).

    44 Country of Publication Code

    Each country name in the UDC placeschedule has been assigned a three-digitednumeric code in an ascending order. The coun-try of publication code is assigned to everyrecord or entry, except the reference entrieswhich are not to be indexed. The tides alongwith the country of publication codes are

    Ann Lib Sci Doc

  • COMPUTERIZATION OF THE NATIONAL UNION CATALOGUE OF SCIENTIFIC SERIALS

    l"ig.2

    Code

    185200185250

    185350

    185501

    185540

    185600

    186800

    186900

    187000

    187100

    187200

    187300

    187400

    187500

    187600

    187700

    187800

    188000

    188100

    188300

    SPONSOR CODESample Data

    Sponsor Heading

    INTERNATIONAL ROAD FEDERATIONINTERNATIONAL RUBBER REGULA-TION COMMI'ITEEINTERNATIONAL RUBBER STUDYGROUPINTERNATIONAL SCHOOL OF SUB-NUCLEAR PHYSICSINTERNATIONAL SCIENTIFIC RADIOUNION. 191620, U03550INTERNATIONAL SEISMOLOGICALCENTREINTERNATIONAL SOCIETY OF ARTAND PSYCHOLOGYINTERNATIONAL SOCIETY OFAUDIOLOGY. S38050INTERNATIONAL SOCIETY OF BIO-CLIMATOLOGY AND BIOMETERO-LOGY.SEE INTERNATIONAL SOCIETYOF BIOMETEOROLOGYINTERNATIONAL SOCIETY OF BIO-METEOROLOGY. 187000INTERNATIONAL SOCIETY OF BLOODTRANSFUSIONINTERNATIONAL SOCIETY OF CAR-DIOLOGYINTERNATIONAL SOCIETY FOR CELLBIOLOGYINTERNATIONAL SOCIETY OF GAS-TROENTEROLOGYINTERNATIONAL SOCIETY OF HEMA-TOLOGYINTERNATIONAL SOCIETY FOR HU-MAN AND ANIMAL MYCOLOGYINTERNATIONAL SOCIETY OF LEA-THER TRADES CHEMISTSEE SOCIETY OF LEATHER TECH-NOLOGISTS AND CHEMISTS (LONDONINTERNATIONAL SOCIETY OF NEPH-ROLOGYINTERNATIONAL SOCIETY FOR PHO-TOGRAMMETRYINTERNATIONAL SOCIETY OF PLAS-TIC SURGEONS. SEE INTERNATIONALCONFEDERATION FOR PLASTIC ANDRECONSTRUCTIVE SURGERY

    Vol 30 No 1 Mar 1983

    188400 INTERNATIONAL SOCIETY OF ROCKMECHANICSINTERNATIONAL SOCIETY OF SUGARCANE TECHNOLOGISTSINTERNATIONAL SOCIETY OFCANE TECHNOLOGISTS, PATHOLOGYSECTIONINTERNATIONAL SOCIETY FOR TER-RAIN VEHICLE SYSTEMSINTERNATIONAL SOCIETY OF URO-LOGY. S38650JNlfERNATIONAL SOLAR ENERGY SO-CIETY. S58650INTERNATIONAL STATISTICAL INSTI-TUTE.136750INTERNATIONAL SUGAR COUNCIL.

    188730

    188750

    188900

    189100

    189150

    189200

    189300

    identified for the generation of the country ofpublication code file. The country of publica-tion code is also used in assigning a subjectheading code which requires further classifi-cation by country. After generation of thecountry of publication index file, the codesare converted by the country names with thehelp of country of publication code dictionaryfile (Fig.4).

    5. INPUT DESIGNFor mechanization of any bibliographic

    record, proper understanding about the prob-lems of the project on the part of the SystemsAnalysis Group is essential. The objectivesand requirements for computerization of theproject are explained to the Systems Analysisgroup of Insdoc in detail. Various types ofproblems connected with the entries are ~~tforward with varied examples, so as to facih-tate the Group to have proper understandingto the intricacies. Based on these problems ,andrequirements, the Group undertook designingthe system compatible to batch mode of IBM360/44 computer. Sample data of about 500entries are coded according to preliminarycard designs and data are key-punched on to80 column punched cards. Before the sampledata are put to test, it has been decided toobtain the output of main part entries andindexes in twenty eight characters per linein three column format. Therefore, at thecoding stage, it is possible to have control

    39

  • Fig.3

    Code

    100200100210100215100230100262100265100400

    101240102280102300

    102360102370

    102380

    102400102600108400110000110200112600

    112800113000

    118000

    118200

    122000

    40

    DUTTA AND OTHERS

    SUBJECT HEADING DICTIONARYSample Dam

    Subject Heading

    5/6 SCIENCE AND TECHNOLOGY5/6:061-6 RESEARCH INSTITUTIONS5/6:82.03 TECHNICAL TRANSLATION5/6 (043) DISSERTATIONS5/6 (091) HISTORY OF SCIENCE5/6 \092) SCIENTISTS BIOGRAPHY5 MATHEMATICS AND NA"tURALSCIENCES502.7 NATURE AND WILD LIFE51+53 MATHEMATICS AND PHYSICS51+531 MATHEMATICS AND MECHA-NICS51/56 PHYSICAL SCIENCES51/54 MATHEMATICS AND CHEMIS-TRY51/53 MATHEMATICS, ASTRONOMYAl'JD PHYSICS.51 MATHEMATICS511 ARITHMETIC512 ALGEBRA512.9 TENSOR513 GEOMETRY513.8 NON-EUCLIDEAN, MULTIDIMEN-SIONAL AND OTHER FORMS OFGEOMETRY513-83 TOPOLOGY514 TRIGONOMETRY. POLYGONO-METRY516 ANALYTICAL AND CO-ORDINATEGEOMETRY516.1/04 CLASSICAL ANALYTICALGEOMETRY517 ANALYSIS. CALCULUS.

    Figo4

    UOCNo.

    (100)(410)(417)(430)(436)(437)(438)(439.1)(44)(45)(46)(469)(480)(481)(485)(489)(492)(493)(494)(495)(497-1)(497.2)(498)(51)(512.317)(519)(520)(540)(548.7)(549)(549.3)

    COUNTRY OF PUBLICATION CODEDICTIONARYSample Data

    Code

    001213264359394405408413414457482497520521522523526532542546553554555560564573577597632635641

    Country

    International OrganizationsGreat ~ritainIrelandGermanyAustriaCzecheslovakiaPolandHungaryFranceItalySpainPortugalFinlandNorwaySwedenDenmarkNetherlandsBelgiumSwitzerlandGreeceYugoslaviaBulgariaRumaniaChinaHong KongKoreaJapanIndiaSri LankaPakistanBangladesh

    Ann Lib Sci Doc

  • COMPUTERIZATION OF THE NATIONAL UNION CATALOGUE OF SCIENTIFIC SERIALS

    over the end of each line either to split or totake the word to the next line. The outcomeof several experiments of the sample dataleaves sufficient scope for further modifica-tions and changes. After extensive experi-mentation and development of software, 15card designs are finalised for conversion of data.

    The structure of different card designs,development of relevant software package forinput and output have been detailed by Murty& Arora (2). Card Design (CD) 41-49 are meantfor coding the main entry and cross referenceentry data. CD 50 is for coding the holdings.In case of CD 42-49, the 80 column punchedcards are divided broadly in three fields. Inthe first field the card columns (CC) 1-22contain all the relevant codes of the entry.The second field CC 23-78 contain the textmatter which is to appear in the publication.In this field CC 50 is the controlling point foreach line in the text portion, so that the mattershould be limited upto this point or before, orlogically split the words into two for furthercontinuation to the next line on CC 5] -78. TheCC 79 has been left blank. The third field CC 80contains the sequence number of continuationcards (0-9). For CD 41-55, CC 1-2 is representedby two digit card design codes, i.e. 42 and 43 fortitles, 4447 for sponsoring body, 48 for imprintand 49 for bibliographical note. Rest of the carddesigns, i.e., 41, 50, 52, 53, 54 and 55 have beendealt with separately.

    For CD 41-50, CC 3-8 contain two digitcoder's code and a four-digited serial number.This helps to retrieve the data from coding sheetfor incorporating necessary correction or dele-tion before the data is key-punched. As such,these codes are not generated on magnetic tapes.The seven-digited alpha-numeric title code iscoded on CC 9-15 for all CD 41-49. Further, incase of CD 44-47 the six-digited alpha-numericsponsor code is coded on CC 16-21 and CC 22left blank. But in remaining card designs exceptCD 41 and 43, CC 16-22 are left blank. Theentries pertaining to each card design from42-49 must contain title code. The other featureelements and specialities of the entry are codedin the summary card (CD 41). The card designsnamely Summary card (CD 41), Holdings data(CD 50), Cross reference to sponsoring body

    Vol 30 No 1 Mar 1983

    index (CD 52), Cross reference entry to indivi-dual library me (CD 55), are explained further.

    6. SUMMARY CARD (CD 41)

    This design incorporates all the salientfeatures of the entry in the coded form. It isprogrammed to check the codes and accord-ingly search for relevant card designs. Like othercard designs, CC 1-2 contain card design number41, CC 3-8 contain two-digited coder's code andfour-digited serial number of the record. CC 9-15 contain the ~le code, on CC 16-18 is codedthe country of publication code, and CC 19-24contain the subject heading code. CC 25 iscoded with 1 or 2 for a main entry or cross re-ference entry. In case of a main entry, CC 27is coded with 1,2 or 3 to indicate presence ofthe sponsor's name with title, sponsor presentseparately, or sponsor completely absent- Thusit helps to identify the sponsor from CD 44 and46. Maximum provision for including twosponsoring bodies in the entry has been made.For me sponsoring body 1 is coded on CC 28,and 2 for two sponsoring bodies. It has alsobeen considered whether the sponsoring bodyappears in the main part has to appear in thesame manner in the index or not- If it appearsin the same manner number 1 is coded on CC 29then CD 44 or CD 44 and 45 are necessary. Ifsponsor appears differently in index, CD 46 orCD 46 and 47 are needed alongwith CD 44 orCD 44 and 45 in case number 2 is coded onCC 29. Accordingly for index purpose CD 30is coded with number 1 for one sponsor andnumber 2 for two sponsors and number 3 forsponsor not present. For presence or absenceof bibliographical note is coded as 1 or 2 onCC 31. Only in case of 1 coded on CC 31,CD 49 will be nceded.CC 34 is coded with 1 or2 in case of a current or a ceased serial. Withthe help of this it is possible to ascertain thetotal number of current titles in the file. Ifnecessary, it may be possible with the help ofCC 34 for further manipulation to generatefiles of current titles only.

    Not all the cross reference entries requireto he indexed. CC 26 is coded with number 1,if the entry is to be indexed. Cross referenceentry, not needed for index, CC 25 is codedwith number 2. Then CC 30 will indicate the

    41

  • DUTrA ....\ND OTF!ERS

    presence or absence of sponsoring body in theindex by coding 1 or 2 for presence of one lhtwo sponsors, and 3 for absence of sponsor.

    The summary card is read while loadingthe data and the required feature elementcodes as well as the related card designs arechecked. If all the necessary items are availablein a proper sequence and position then therecord is loaded on magnetic tape. Otherwise,if anything is found incorrect or not in appro-priate position then the data relating to thatparticular record is not loaded. •.

    7. HOLDINGS INFORMATION (CD 50)

    This desicn is different from the main entry,~card designs (42-49). Representation of bothtitle code and library symbol are absolutelynecessary while coding the holdings data of eachlibrary. A six-digited library symbol consistingof capital letters of Romal alphabet has beendesigned for earn participating library. Foroperating the related programmes, the 80column punched card is divided into sevenfields. First field, CC 1-2 is coded by card designnumber 50. Second field, CC 3-8 contains two-digited coder's code and four-digited serial num-ber for each library holdings information. Thirdfield, CC 9-14 is coded with a six-digited alphacode which also represents the library symbol.Fourth field, CC 15-21 contains the title codeof the corresponding holdings information,Fifth field, CC 22 is either coded by 1,2 or 3 toindicate different types of holdings. Number 1indicates the complete set without any missingor incorplete volume, number 2 indicate thepartial holdings, i,e, holdings right from begin-ning or from intermediate point, but havingmissing or incomplete volumes, and number 4indicates stray holdings having only one or twovolumes or only a few issues of a serial.This typeof catagorization in holdings is made with aview to print only complete holdings or bothcomplete and partial holdings with the mainentry, if needed. Sixth field, CC 23-62 containthe holdings data. CC 42 and 62 are the control-ling points for logically breaking the holdingsdata or further continuation to succeeding line.Seventh field, ex: 80 is coded by 0 to 9 as thecase may be to govern the sequence of cards foreach holdings information.

    42

    8. CROSS REFERENCE TO SPONSORINGBODY INDEX (CD 52)

    While generating :he sponsor index file, thesponsoring body headings along with the relatedtitles arc selected from the entry file. The spon-SOl ng body headings an: shown under the latestki. wn names or the vernacular fonns of names.Tl: Ie is no ~~opc in the system to incorporatec: ,,_,:; reference headings from earlier knownnames or the variant names in the sponsoringbody headings file. This form of cross reference~leadings is provided externally in the system.Evidently CD 52 has been designed for this pur-pose. After ascertaining the cross reference head-ings required for the file, the data is coded. Themain fields in the design are CCjl-2 contain thecard design (CD 52). CC 3-i contain coder'scode and a four-degited serial number for therecord. On CC 9-14 are left blank. CC 23-78contain the sponsoring body headings as itshould appear in the index, and CC 79 is keptblank. As usual CC 80 has been kept for deter-mining the sequence of continuation cards (0-9).

    There are two more card designs meant forgenerating subject heading dictionary (CD 53)and country of publication code dictionary (CD54) respectively. The structure and function ofthese designs are simple. Coding of subject head-ings dictionary (CD 53) is done with card desi~number 53 on CC 1-2. On CC 3-8 contamcoder's code and new serial number. On Cr::9-14 is coded by a six-digited subject code, andCC 15-22 are left blank. CC 23-78 are codedwith UDC number and corresponding subjectheading, leaving a space blank in between twoitems. In case of country of publication codedictionary, CC 1-2 contains the card designnumber 54. On CC 3-8 is coded coder's codeand new serial number, and on CC 9-11 is codedwith country of publication code. CC 12-22 areleft blank, and CC 23-78 contains the countryof publication code with name of the place/country. As usual, for both the above mentione?designs theCC 79 is left blank and CC 80 IScoded to indicate the sequence of continuationcards (0-9).

    Ann Lib Sci Doc

  • COMfUTERIZATION OF THE NATIONAL UNION CATALOGUE OF SCIENTIFIC SERIALS

    9. CROSS-REFERENCE ENTRY TO INDI-VIDUAL LIBRARY HOLDINGS FILE(CD 55)It IS planned to bring out catalogues of

    individual participating libraries from entry fileand holdings file. It has already been mentionedthat holdings file is generated as a separate unit,but information coded for each library's hold-ings is always accompanied by the correspondingtitle code. The data base of the entries of serialsis matched with the holdings file. The holdingsfile is firs; sorted by title-code-wise and then bylibrary-symbol-wise. This matching of two filesgenerates the main part for a union catalogue.To bring out catalogue of individual library,the holdings file is resorted first by library-symbol-wise and then by title-code-wise. Thenew file consists of main entries with corres-ponding library holdings. There will be seriouslacunae, if earlier title held by the library andthe alternate title reference are not accountedfor along with the main entries. The provisionof earlier title or alternative title reference willinvariably vary from one library catalogue toother depending upon the holdings information.At the time of consolidating the library hold-ings information, it is given considerable impor-tance to indicate the cross reference title code

    Vol 30 No 1 Mar 1983

    against the concerned library symbol in themain entry slip, if the library holdings hascovered for earlier title. Therefore, this carddesign is meant for coding the data of crossreference entries required for individual librarycatalogue. CC 1-2 consist of card design number5~, and CC 3-8 contain two-digited coder's codeand four-digited serial number. On CC 9-14 iscoded with the library symbol, and on CC 15-21is coded with cross reference title code. CC 22is coded with number 4 to differentiate it fromother types of holdings information. The addi-tional data of cross reference title codes areincorporated in the holdings file. The file issorted first by library symbol-wise and then bytitle-code-wise. The file now generated containsthe codes for main entries and cross referenceentries in a sequence under each library symbol.

    REFERENCES

    1 Porter, K.I. The New Bucop: the problemsof the documentation of serial publicationsASLIB Proceedings. 1964, 16(6),189-199.

    2. Murty D.S.R. & Arora, A.M. Processing ofUnion Catalogue of Serials data using IBMsystemf360 computer. Ann Lib Sci Doc,1974,21(3),85-94.

    43