PROJECT-ORIENTED DOCUMENTATION:A NEW APPROACH TO BIBLIOGRAPHIC SERVICE

The documentation service a8 origil)ally introducedin the National Metallurgical Laboratory, the dr-aw-backs of this system and the details of a revised newapproach to documentation which was subsequentlyevolved are discussed. The procedure followed forthe collection, processing and use of t he informationunder the new set-up, comparative •••advantages of thenew approach over the previoua method and problemafaced in. the development of the new system are pre-sented.

Introduction

The importance and appropriateness ofdocumentation and information service wasrealized in the National Metallurgical Labora-tory quite early in its life. Among the variousinstitutions of its kind that carne into existencein the wake of national Independence, this wasone, whose efforts were very early tested inpractice, and on whose skills and ingenuitythere has been a growing demand as a verynatural direct consequence of the industrialexplosion through which post-independenceIndia has been passing. It was found thatmetallurgical information appearing in the mostadvanced journals of the West was only partlyapplicable to its industry and technology, aswas also the information obtainable in the booksWhile the former was far ahead of its require-ments, and had little relevance to its presentconditions, the latter was not advanced enoughfor its needs, or did not provide all theanswers to the numerous problems of itsadaptation to the technological, economic andsocial environs of this country.

The Enlphasis on PurposeAs a result, from the beginning, a

search was made for a method which wouldensure that documentation service, which alonecould bridge the information gap, be selectiveand purposeful in that it must meet the specificneeds of projects in progress. In other words,it must cater to a ~emand for technologicalinformation which the industry waited for trans-1ating into practice for the advancement of thecountry's economy.

Vol 20 No 1-4 Mar-Dec 1973

Kalyani Banerjee

National Metallurgical Laboratory,Jarnshedpur.

In this attempt, there were two difficul-ties. First, no matter how qualified or percep-tive the documentation staff could be, theycould not always evaluate or assess the suita-bility of a particular information to the specificdemands of a case. Second, even if, somehow,the necessary items could be selected, therewas the difficulty of reaching it to the appro-priate persons on account of the production'bottlenecks and mechanics of distribution.Because of the system of production in vogue,viz , , typescript sheets of foolscap size, onlyfour or five copies could be produced and theycould either fail to reach all the scientists whoare interested to receive it, or reached themtoo late in the line of routing to be of muchinterest or use to them.

It may be argued that this could be solvedby the process of rnaktng multiple copiesthrough cyclostyling. But, even then, as thematters of interest to a particular division wasscattered in a large volume of material goingup to over 40 pages, it was not easy for thatdivision to locate the items of interest quickly.It would be no use here to advance the view thatthey could be taught the classification sv st em,so that they could at least locate their items byfollowing the classified order. But, in a rnult i-faceted close classification, that has to be usedin abstracts and documentation services of thistype, any reader would be completely lostunless he was trained in l ibr ar ianship , Besides,there was no way of leaving with them thedesired material for reference and use. Thisdifficulty became apparent early in the experi-mentation with documentation.

Initially, the documentation service inN. M. L. followed the pattern adopted, else-where, and had the same limitations of selec-tion, production and uti! ity.

63

BANERJEE

Selection of Items

The material was selected by the librarystaff from the periodicals received in the Lib-rary on the basis of the best judgement thatcould be formed about the needs of the staff.particul arl y the research and scientific staff.But as the Library had no living contact withthem. it was not possible exactly to foreseetheir needs. This was also true of the Liaisonand Information Division of the Laboratory. Ithad. no doubt. a much closer contact with themainstream of industry, but it .would seldomthink of taking the library into confidence. oruse its resources for meeting its requirementsor suppl ement ing its solutions to the problemsthat the industry referred to it. Most of thetime. it prepared its answers on the basis ofthe work done or from the opinions of thespecialist in the fields. without consideringthat the library and its information servicescould be profitably used to modify or to supple-ment them. sometimes even radically.

Production Method

The production and presentation was inthat usual booklet form, arranged into certaingroups. Since it was produced as a typescript.the copies were limited to six at the most and.as such. could not be circulated to all. Thedocumentation slips. on the bas is of which itwas compiled. were also limited to six copiesor even less since. for preservation. they hadto be printed on thicker paper. The circulation,thus, was extremely limited. But even if itcould be enlarged. the reader sometimes feltthat:

He had no need for the items included.and missed items that were needed byhimj ,

2 Even if it contained some materialwhich was useful to him. he had towade through a large volume ofus el es s material to find it.

As SUCh. he felt rather unenthusiasticabout this compilation.

Extent of Use

The util ity of this proj ect was thus of avery restricted nature. Not so much becauseof its intrinsic lack of worth. but because ofthe selection and production methods. Evenotherwise, it could not pinpoint the needs of

64

the user and cater to them. This lack ofutility was most frustrating. when it wasviewed against the tremendous effort spent inthis fruitles s pursuit.

Project-Oriented Documentation

To obviate all these difficulties. a newsystem of Proj ect -Or-iented Documentation(POD) was conceived. and put into operation in1964. This was started in the National Metal-lurgical Laboratory as a pilot project. underthe CSIR. The Indian National Scientific Docu-mentation Centre also was associated with itand took a keen interest in this project. andrendered valuable suggestions and guidance inits successful development. The essentialfeatures of the project are:

Controlled circulation and decentra-lized screening by scientists.

2 A new mechanics of production.

3 Decentralized filing and more exten-sive distribution.

Circulation and Screening of Journals "

CirculationIt may not be out of place here to mention

some of the problems associated with the cir-culation of journals among the scientific staffwhich have been successfully tackled in thecourse of the development of this system. Inthe beginning. the circulation of journalsproved a great bottleneck, when the routingwas from one department to other, the circuttstarting with the Library and ending in it.Under this system. divisions down the linereceived the journal too late. and the Librarycould never be too sure which division had ajournal. They also seldom came back to theLibrary in time. As such. many items couldnot be incorporated in the abstracts at theappropriate time.

This problem was solved by changing therouting system and asking each division toreturn the journals sent to it within 48 hoursof rec eipt , so that (a) the Library could keeptrack of it and (b) each division got it directfrom the Library and returned to it. By thischanged procedure, the control over the move-ment of journals became much more effective.

However. the changed method as well asthe former procedure left one problem unsolved,

Ann Lib Sci Doc

PROJECT ORIENTED DOCUMENTATION

viz., the loss of Journal during circulation,which could occur either in transit or while inuse in the receiving division. An adequate yetsimple procedure was evolved to prevent suchlosses in circulation. Individual issue regis-ters containing the following entries are usedfor this purpose:

Date of despatch to the division/section,

2 Code and date of publication of thejournal,

3 Signature of receiver.

4 Date of its return to the Library fromthe receiving division.

The second point needs a little explana-tion. To avoid writing the long names of thejournals each time in the peon book. all jour-nal s were given a numerical code number.which roughly followed an alphabetical order.This looked something like. say 648/71/5/27.meaning thereby the issue of 1971, 27th May.of the Journal No. 648. In giving the codenumber to the journals in the alphabeticalorder, enough gap was left for future insertionso that the codes did not have to be revised 'too frequently. and could acquire some mnemo-n~c .v~lue both to the Library and the receivingdtv is ion s , thus helping quick references. Thisprocedure has simplified the work considerably,and has ensured a practically faultless circu-lation system.

Screening by Research Staff

For this purpose, the journals and otherperiodical literature received in the Libraryas also patent specifications and reprints are,after 7 days display in the Library, groupedaccording to the functional divisions of theLaboratory and the arrivals in the relevantgroup sent to the concerned divisions for cir-c~~t~on. The Laboratory has the following 19dtv is ion s among which the journals are circu-lated.

Director

2 Director's Laboratory

3 Deputy Director

4 Dy. Director's Laboratory

5 Alloy Steel Division

6 Chemistry Division

Vol 20 No 1-4 Mar-Dec 1973

7 Corrosion Section

8 Creep Laboratory

9 Design & Engg. Section

10 Electrical Section

II Electronics Section

12 Extractive Metallurgy Di.vision

13 FPT Division

14 Genl. Metallurgy Division

15 Mechanical Testing Division

16 Ore Dressing Division

17 Physical Metallurgy Divi.sion

18 Refractory Division

19 Research &. Covo r dtnet ton

Each division, on receipt of the j ou rnaI acoming to it, is required to go through th e rr-and mark the items needed in pencil, rr.e nt io n-ing the page numbers of the marked items O~

the routing slip in pencil. and return the, _!('Ul·-nals back to the Library within 48 hou r s . illmaking the selections, they are r equ es te d to to""-wide, yet discriminating. They a r e told ;:,:,select items which are of irn rn ed iat c as well :c.,"possible future int ••rest --the p r inc ip al. cr ite risbeing their applicability or r el evance to theassignments given to them. They can s el.e c tany published item, be it an article, note, bookreview and even advertisement wh ic h n.i av ~.":-some use.

After the journals COTneback to the Li.t,-rary, the Library Officer goes through the rnonce again and ma y delete it e rn s that ;rc rep',-titive or add those that have been missed. Therepetition comes mostly from those itemswhich s imul tan eou sl y b r eak out in s evc t: aljournals in the same dt sc ipl ine. h may be .:0

article that has been reprinted in s?ve;aljournals, or a news it e m or an e xt r a.ct , '"'f}'l1.3

is done to avoid unnecessary dupl ic at ioc . V.such cases, the original source or the be,,,::·presentation or production is traced and <''''.'.:this retained in preference to the others. 'n-: ..-repetition arises out of the fact that 2_E tL,·Library sub s c rib e s to many journals j,. r'T",·:;·pl e copies and they go to all the di vis con" ,:hesame item may often be selected by rno r e t\""one division. .

BANERJEE

By this process of screening and selec-tions, three points are ensured:

That the screening becomes moreexhaustive as the work is divided,onlya few journals are handled by a per-son, each journal going to one spe-cialist mostly. These are normallyscreened by the second level scientificstaff, who are in many cases of a postgraduate qualification, such as Sr.Scientific Officer, etc. They are notonly fully qualified to be aware of theimportance of an information, butbeing immediately in touch with thescientist in charge, as well as in touchwith experimental work, they are well-qualified to make an assessment ofthe value of the selected item.

2 Such preliminary screening ensuresthat only items of actual interest willfind its way into the documentationsystem, thus establishing a firm basisfor it s util ity.

3 The process of screening itselfattracts the interest of the scientificstaff to the new developments and pro-gress in that area and they get aninvolvement in the quality of the docu-mentation service, which graduallybecomes their own work rather thanthe Library's. This has a good effecton their use of, and relations with,the Library also.

4 Since every item is screened by aspecialist, and selected by him, thedocumentation staff can also drawupon their help while classifying theseitems, particularly, when it is beyondthe limits of their acquaintance withthe subject, or it is a new subject ora very narrow special ization. In suchcases, a reference to the screeningstaff becomes the easiest way ofcorrectly classifying the subject. Thiscan be done by a simple telephone callon the internal telephone. Formerly,such help could be obtained only bysending the papers to the specialist,or even finding who the specialistconcerned would be, and requestinghim to read the item and give anopinion.

66

Production and Classification

Editing and Collation

The journals received back after screen-ing are collated and the marked material edited,by deleting portions or marking points of bib-liographical importance, so that the typistcould type them in the standard format andstyle and in standard size slips in four to sixcopies - four for items which are wanted byonly one division and six for items where twoor three divisions are involved. If required,more copies are also made, but that involvesdouble typing. Such cases are very few. Afterthe slips are typed, they are compared by thestaff of the Library. An attempt is being madecurrently to persuade the screening staff tocompare thern for accuracy, because wheremathematical symbols or scientific jargon isus ed, the typists or the stenographers maysometimes mi ss a vital figure or a symbol ormisspell a word, completely twisting themeaning or making the abstract useless. Forthis purpose, the scientists selecting andscreening them are being requested to putidentifying marks or initials, so that this pro-cedure could be made a regular feature of thedocumentation process. This will also reducethe load of a brave but foredoomed attempt onthe part of the Library to ensure accuracy inthe documentation system.

Class ification

The clas sification system follows themodified U. D. C. The subject classification andsynthesis is typed on the top right hand corner,about on inch and a half away from the rightedge in one or more lines, depending on thelength of the notation. Generally, the lineationis according to the component fac ets of the sub-ject. As already stated, the work of classifica-tion is greatly aided by the research staff of theLaboratory.

Production

In this connection, a problem of repro-duction that has been besetting the Library maybe mentioned for discussion and a solution.The slips for the index file are of a stockwhich has to be thick enough to afford verticalfiling and manipulation. Such thick papercannot take four or six copies. Even withelectric typewriter, because of the thicknessof the paper, the fourth or sixth copy becomes

Ann Lib Sci Doc

PROJECT ORIENTED DOCUMENTATION

blunt, due to the edges of the sharp type facestravelling through a thick blanket of paper. So,even though the fourth or sixth copy may havea deep carbon impression, the edges of thetype become blunt almost to the point of makingthem solid, or in any case much bolder thanthey are, so that reading becomes difficult.Cyclostyling also affords no solution, as it hasto be done on the standard paper suppl ied bythe manufactur er s , which is too thin for verti-cal filing, and which have t.e be cut to standardsize.

One solution can be to ask the suppliersof cyclostyl Ing paper for a thicker variety,with standard size long perfo r at ion at the sizelimit of the card. Or else, some sort ofreproduction process, which must be cheap andwhich will give good reproduction on any paper,should be used. The use of a Xerox type pro-cess with standard size blanks of the appro-priate stock of plain paper, would probably bethe ideal solution. It is hoped some of thenewer methods which have been recently adver-tised, for instance, Majox by Mcneil &: Barryand the one by Advani-Oerlikon, both based onthe electrostatic principle developed in theNational Physical Laboratory, will be a solu-tion to the problern,

Such a solution will also largely simplifythe work of classification. The cards areclassified after they are typed. The classifi-cation is indicated in pencil in one slip andthen typed in the others, since it has not beenpossible to develop an aesthetic hand-writingof uniform size to accommodate the entirenotation. Such typing also involves at leasttwo insertions and much typing work. Eventhen, the impression is not uniform as theregistration of the cards are seldom perfect.

Distribution

The documentation slips are arranged intheir classified order and made into setscovering the subjects under a division.

The first set of all slips, which go tothe Library central file, is interf iledwith the slips already there.

2 The copies meant for the divisions aregrouped together and arranged in theorder of classification, bound withclip fasteners, and each division issupplied with one bound and one loose

Vol 20 No 1-4 Mar-Dec 1973

set _ the latter for interfiling in thedivis ion file.

3 The Director's set is bound with cordas it becomes too thick for the clipfasteners. All sets for the Directorare cut to trim the edges to a finefinish.

Review and Evaluation

The' system has been operating for overseven years now. Some of the problems thatwere faced and overcome in course of itsoperation may be mentioned in brief.

Early in the process, it was found thatthe readers received the journals at least 10-20 days after they were put in the chain ofcirculation. The damage due to this delay hasbeen minimized by is suing a title index of newarrivals (Index of Current Titles) every week.This Index keeps the scientists concerned awareof the arrivals and they even mark in their owncopy the items that they would like to read/select. This contains only the title and authorof the main articles arranged according to thetitle of the journal. This is produced by cyclo-styling and distributed to all senior scientificstaff.

This serv ic c has been very much appre-ciated and has added to the utility and improve-ment of the system.

Author index

A suggestion was received from some of,the users to introduce a feature in the systemwhich is in use in many places, viz , , anauthor index to the documentation scheme, tobe issued monthly for the items compiled in amonth, and accumulated at the end of the year.It was argued that such an index would

Help the research workers, who veryoften refer to published information interms of their authors;

2 Bring into focus the published work ofthose whose research and experimen-tations break new grounds or consti-tute the definitive work in any disci-pline;

3 Give an idea of the standard of theitems abstracted as reflected in thereputation of their authors.

67

BANERJEE

Even while agreeing with the argumentsadvanced. it was not felt necessary to introducethis feature. because it was found that if theauthor of a publication was known. it would beeasily traced from several other abstractingand indexing systems. such as ISI cards. andthis would answer the first two points. Asregards the third. though partly true the exer-cise was found to be too strenuous to be justi-fied by this doubtful standard for assessmp.nt.First, many reputable authors' published work.which even though of a first class standard.had no relevance in our context. Second, theauthors who were comparatively less knownconstituted the bulk of the contributors in anymonth's compilation. This suggestion was.therefore, abandoned.

Merits of the new system

1. Greater use

The new system has found a far greateruse than the preceding one. Though no statis-tics has been maintained, each division con-sults its index very frequently. They have bynow collected a sizeable body of information ofdirect relevance to their own work, and, forthis, has set up their own 'mini-documentation'unit. Their keenness to have it is also appa-rent from the fact that if a division does notreceive its share of the compilation any monthwhen somebody else receives it, they imme-diatelyenquire about it. Even otherwise, theyenquire about its pos sible publication date.This shows that there is an expressed demandfor it. It was not so with the older system.In the past, there was no way of knowing evenby this indirect method of demand. how'muchuse was made of the system or its utility. Thisuse, as already mentioned has had a multipliereffect. Arising out of this, scientific staffoft en want supporting material, particularly inthe area of patents, specifications, drawings,etc. which are a direct result of this service.

2. Compilation of bibliography

As a result of the documentation service.the compilation of bibliography has not onlybecome considerably easy. it has alsoimproved in quality and quantity. Even on thenarrowest specialization, the Library can nowcompile a pin-pointed bibliography in any areawithin the specialization of the Laboratory, inpractically no time, covering a wide field ofas many as 30-40 items. This service has

68

been universally appreciated not only by thestaff of the Laboratory, but by others in thefield who have had occasions to draw on itsresources.

3. Proj ect planning

Scientists now extensively use the docu-mentation system of the Library while pre-paring research projects and plans of work to'be undertaken. As a result of the use of thisdocumentation system, their plans becomemuch more precise and practical, and litera-ture search, which forms the first step in anyresearch project, now poses no difficulty of.search or time as before.

Conclusion

1n conclusion, a point about the econo-mics and staffing requirements of this servicemay be of some interest. The service involvesthe handling of 600 per iodical s of all types in amonth running into a total of 900 in terms ofsingle publications. The patents account forapproximately another 100 items per month.

The document"tion slips number approxi-mately 500 items in any month. The circula-tion and distribution involve 19 points. Theentire work is carried out with a staff of onlytwo people: the author, in addition to her workas in-charge of the NML Library, and onetypist. While starting the project, it wasthought that a staff of 5 documentalists and 3stenos would be required for this job. Althoughthis never materialized, the service has beenmaintained by the dedicated and hard work onthe part of the staff conc erne d, only because itmet an expressed demand of the readers. Onlylibrarians will understand the extent to 'whichprofes sional pride and the gratification athaving metafelt need of readers will motivate alibrarian.

Acknowledgement

The author thanks Dr. V. A. Altekar,Director, National Metallurgical Laboratory,for his encouragement and guidance in the pre-paration of this paper. and his kind permissionto publish it.

Ann Lib Sci Doc