College and Research Libraries B y I . A . W A R H E I T A Study of Coordinate Indexing as Applied to U.S. Atomic Energy Commission Reports Dr. Warheit is head, library department, Argonne National Laboratory, Lemont, III. SHORTLY AFTER it had been proposed that the U n i t e r m system of C o o r d i n a t e In- d e x i n g be adopted f o r the research reports of the U . S. A t o m i c E n e r g y C o m m i s s i o n , some of the i n f o r m a t i o n officers and l i b r a r - ians concerned w i t h the u t i l i z a t i o n of A E C research reports u n d e r t o o k a series of inde- pendent studies of both a theoretical and p r a c t i c a l n a t u r e . U p o n comparison of the results of these studies the i n f o r m a t i o n officers and l i b r a r i a n s concerned agreed that a g r e a t m a n y p r o b l e m s remained to be resolved, t h a t the system of c o o r d i n a t e in- d e x i n g requires f u r t h e r d e v e l o p m e n t and that in its present state of d e v e l o p m e n t , the c o o r d i n a t e i n d e x i n g system o f f e r s no induce- m e n t f o r c h a n g e , a n d , indeed, its applica- tion to A E C reports w o u l d be a r e t r o g r a d e step. I t is the purpose of this paper to s u m m a r - ize the m o r e p e r t i n e n t of these tests t h a t w e r e applied to the U n i t e r m S y s t e m of C o - o r d i n a t e I n d e x i n g , o m i t t i n g , as f a r as prac- tical, the v a r i o u s t h e o r e t i c a l studies. 1 A l - t h o u g h the results w e r e n e g a t i v e in the case of the A E C reports, it does n o t f o l l o w t h a t the results w i l l be the same f o r a n o t h e r body of l i t e r a t u r e . I t is o n l y a f t e r a series 1 Jack C . M o r r i s , " E v o l u t i o n or I n v o l u t i o n ? " Journal of Cataloging and Classification, X ( J u l y , 1 9 5 4 ) , 1 1 1 - 1 8 . T h i s w a s part of the A E C study. See also his paper " T h e D u a l i t y Concept in S u b j e c t A n a l y s i s , " American Documentation, V ( A u g u s t , 1 9 5 4 ) , 117-46. See espe- cially pp. 138-146. G . E . R a n d a l l , " P r a c t i c a l i t y of Co- o r d i n a t e I n d e x i n g , " C O L L E G E A N D R E S E A R C H L I B R A R I E S , X V (October, 1 9 5 4 ) , 4 1 7 - 1 9 . of tests are applied to v a r i o u s collections 2 t h a t a definite e v a l u a t i o n of C o o r d i n a t e In- d e x i n g as a w h o l e can be m a d e w i t h real assurance. A l a r g e b o d y of e x p e r i m e n t a l d a t a is necessary in o r d e r to reduce m u c h of the c o n t r o v e r s y n o w r a g i n g about this n o v e l f o r m of i n d e x i n g . O n e of the m a n y a d v a n t a g e s claimed by the a d v o c a t e s of C o o r d i n a t e I n d e x i n g is t h a t the card c a t a l o g is r a d i c a l l y reduced in size and t h a t there is a s i m i l a r r e d u c t i o n in the l i b r a r i a n ' s w o r k l o a d . 3 A comparison there- f o r e w a s m a d e b e t w e e n an e x i s t i n g A E C reports c a r d c a t a l o g f o r 40,000 d o c u m e n t s and its e q u i v a l e n t as a U n i t e r m c a t a l o g . Present Uniterm Catalog Catalog 1. Series ( N u m b e r ) C a r d s 55,000 55,000 2. A u t h o r C a r d s 102,000 102,000 3. Accession N u m b e r C a r d s o 40,000 4. Subject C a r d s 180,000 6,000 337,000 203,000 I t should be pointed o u t t h a t the f i g u r e of 6000 U n i t e r m s u b j e c t c a r d s is based on c l a i m s m a d e by D o c u m e n t a t i o n I n c . as a result of their analysis of the s u b j e c t head- ings used by the T e c h n i c a l I n f o r m a t i o n D i v i s i o n ( T I D ) of the L i b r a r y of C o n g r e s s - F o r another test of Coordinate I n d e x i n g see C . W . C l e v e r d o n and R. G . T h o r n e . A Brief Experiment with the Uniterm System of Co-ordinate Indexing for the Cataloging of Structural Data. ( R o y a l A i r c r a f t Estab- lishment, L i b r a r y M e m o r a n d u m No. 7) J a n u a r y , 1954. 3 C . D . Gull, " A l p h a b e t i c S u b j e c t I n d e x e s and Co- ordinate I n d e x e s : a n E x p e r i m e n t a l C o m p a r i s o n , " COL- L E G E A N D R E S E A R C H L I B R A R I E S , X I V ( J u l y , 1 9 5 3 ) , 2 7 6 - 8 1 ; also as chapter 6, p. 56 of M o r t i m e r T a u b e & A s s o c i - ates, Studies in Coordinate Indexing. ( W a s h i n g t o n : Doc- umentation, I n c . , 1 9 5 3 ) . 278 COLLEGE AND RESEARCH LIBRARIES and of the D o c u m e n t S e r v i c e C e n t e r ( D S C ) in D a y t o n . 4 A c a r e f u l c o u n t of the A E C s u b j e c t h e a d i n g list s h o w e d t h a t it w o u l d take almost 8500 U n i t e r m s to c o v e r all the s u b j e c t s a d e q u a t e l y . 5 B u t in order to reduce the areas of c o n t r o v e r s y , D o c u - m e n t a t i o n I n c . f i g u r e s are used w h e r e v e r possible. A l s o , in this instance, the d i f f e r - ence of a f e w thousand cards w i l l not mate- r i a l l y a f f e c t the results. T h e s e 6000 U n i t e r m c a r d s w o u l d c a r r y 342,000 postings. T h i s is based on 8.55 U n i t e r m s per title. A l t h o u g h G u l l claims " t h a t the a v e r a g e of 6.88 u n i t terms per report is the o p t i m u m f o r C o o r d i n a t e In- d e x i n g of . . . T I D r e p o r t s , " the f i g u r e of 8.55 U n i t e r m s per title is based on a t a b u l a - tion of 3 1 1 A S T I A cards w h i c h c a r r y U n i - t e r m t r a c i n g s . A c t u a l l y the n u m b e r of postings w o u l d be h i g h e r in an A E C cata- l o g . J u s t as there are about t w i c e as m a n y s u b j e c t t r a c i n g s on an A E C c a t a l o g c a r d as compared to an A S T I A card, the coordi- nate i n d e x i n g of one sample of 200 A E C reports required 1 1 . 4 1 U n i t e r m s per title. T h u s one should expect some 4 5 6 , 4 0 0 post- ings, but a g a i n the l o w e r D o c u m e n t a t i o n I n c . f i g u r e s w i l l be used. O n the basis of past experience and a series of test postings, it w a s determined that, day-in, day-out, the a v e r a g e c l e r k c o u l d sort, m a r k and file at the rate of 60 cards an h o u r and hand post at a rate of 40 accession n u m b e r s an hour. T h e p o s t i n g rate m a y seem rather l o w , but it m u s t be r e m e m b e r e d t h a t U n i t e r m t r a c i n g s c a n n o t be presorted b u t m u s t be posted in accession 4 R i c h a r d B . T h o m a s and C. D . Gull. The Choice of Uniterms for a Coordinate Index of Scientific and Technical Reports, T e c h n i c a l Report no. 6 . . . f o r the A r m e d S e r v i c e s T e c h n i c a l I n f o r m a t i o n A g e n c y , M a r c h , 1 9 5 3 ; also as chapter 5, p. 47, T a u b e , op. cit. Combin- i n g the T I D and D S C s u b j e c t heading lists, Documenta- tion I n c . converted them into a list of 6582 U n i t e r m s . s T h e A ' E C subject h e a d i n g list has some 10,000 sub- j e c t headings and a little over 1500 see r e f e r e n c e s . T h e 8500 U n i t e r m s would include 1 1 1 7 names of organic compounds and 1406 proper geographic names used for geology reports. T h e total actual reduction in subject headings if U n i t e r m s w e r e used would be in the order of 4 0 % . T h e figure will v a r y f r o m 3 6 % to 5 3 % depending on the inclusion or exclusion of the geology and organic compounds headings. o r d e r . 6 T h e r e f o r e the w h o l e index m u s t be w o r k e d t h r o u g h f o r each title separately. If m o r e t h a n one posting c l e r k is necessary then a certain type of p r e s o r t i n g is possible. T h e entries c o u l d be g r o u p e d by their last digit and then posted. T h i s w o u l d still mean w o r k i n g t h r o u g h the alphabet f o r each e n t r y b u t w o u l d permit m o r e t h a n one person to do the w o r k at the same time. So, even a l l o w i n g f o r the d e v e l o p m e n t of me- chanical devices to speed the p u l l i n g of c a r d s and the posting of n u m b e r s , it is h a r d l y possible, over a l o n g period of time, to aver- age m u c h m o r e t h a n 4 0 postings per h o u r . O n this basis, ( t h e A E C c a t a l o g w o u l d re- quire 5 6 1 6 hours of w o r k , but the U n i t e r m c a t a l o g w o u l d require 3383 hours f o r f i l i n g and m a r k i n g plus 8 5 5 0 h o u r s of posting, f o r a g r a n d t o t a l of 1 1 , 9 0 0 hours. R e d u c i n g these to more u n d e r s t a n d a b l e figures, w e find t h a t , f o r e v e r y 100 titles received, the regular A E C catalog requires: 843 cards = 14.05 hours O postings = o hours 14.05 hours the Uniterm catalog requires: 508 cards = 8.45 hours 855 postings = 21.37 hours 29.82 hours T h u s , c o n v e r t i n g A E C libraries to C o - o r d i n a t e I n d e x i n g w o u l d mean at least a d o u b l i n g in c a t a l o g m a i n t e n a n c e costs. J u s t to break even it w o u l d be necessary to p u l l , post, and refile 855 items in 5 h o u r s and 36 minutes, or o v e r 1 5 0 items an h o u r ! I t should also be pointed o u t t h a t it w o u l d be v e r y difficult f o r m o r e than one person at a time to w o r k at a U n i t e r m c a t a l o g . Since 6 U s i n g high speed B u r r o u g h s or National Cash Regis- ter machine, the a v e r a g e posting clerk in a bank can post f r o m 400 to 600 accounts each day. T h i s includes the necessary preparation work such as alphabetization of checks and deposit slips, but it does not include verifi- cation of items. JULY, 1955 279 p o s t i n g w o u l d be a f u l l - t i m e j o b , it w o u l d be necessary either to h a v e t w o U n i t e r m s u b j e c t c a t a l o g s , each of w h i c h c o u l d be posted on a l t e r n a t e days w i t h one thus a l w a y s a v a i l a b l e f o r the public, or to h a v e the p o s t i n g c l e r k s w o r k n i g h t s . O n the o t h e r h a n d the f a c t t h a t the C o o r d i n a t e I n d e x requires f e w e r cards, does m e a n a s l i g h t r e d u c t i o n in p r i n t i n g costs. M o r e i m p o r t a n t t h a n cost, h o w e v e r , is the efficiency of the system. C a n C o o r d i - nate I n d e x i n g g i v e b e t t e r r e t r i e v a l t h a n the r e g u l a r c a t a l o g ? T o test this, three l i b r a r - ians at an A E C l a b o r a t o r y U n i t e r m indexed some 500 of their reports in a c c o r d a n c e w i t h the i n s t r u c t i o n s in the Installation Manual for the Uniterm System of Coordinate In- dexing prepared by D o c u m e n t a t i o n I n c . and published by the D o c u m e n t S e r v i c e C e n t e r of the A r m e d S e r v i c e s T e c h n i c a l I n f o r m a - tion A g e n c y , O c t o b e r , 1 9 5 3 . T h e report n u m b e r s , used in lieu of accession n u m b e r s in o r d e r to s i m p l i f y the task, w e r e then posted to the a p p r o p r i a t e U n i t e r m cards. T h e p o s t i n g w a s done by h a n d by t w o people, one p u l l i n g and r e f i l i n g the cards, and the o t h e r posting. I t soon t u r n e d o u t t h a t the t w o p o s t i n g c l e r k s w e r e u n a b l e to keep up w i t h the three indexers. T h e r e - f o r e , w h i l e some 500 reports w e r e a c t u a l l y i n d e x e d , o n l y 200 w e r e e v e r posted. A s a result, the test w a s confined to the 200 titles w h i c h c o u l d be posted. W h e n the sample c a t a l o g w a s c o m p l e t e d , o t h e r l i b r a r i a n s and " s c i e n t i s t u s e r s " experi- m e n t e d w i t h it. A l l w e r e sufficiently w e l l a c q u a i n t e d w i t h the s u b j e c t m a t t e r of the reports i n c l u d e d in the sample to select r e f - erence questions capable of a n s w e r by the m a t e r i a l i n d e x e d . A l m o s t as soon as the tests began, f o u r t h i n g s became a p p a r e n t . F i r s t , there w e r e an i n o r d i n a t e l y l a r g e n u m b e r of false drops. Second, m a n y items c o u l d not be r e t r i e v e d . T h i r d , the absence of a n y descriptive i n f o r m a t i o n or a b s t r a c t m a d e m a n y of the searchers feel they w e r e h u n t i n g blind and b r o u g h t i m m e d i a t e pro- tests. F o u r t h , c e r t a i n key U n i t e r m s w e r e required in a l a r g e p e r c e n t a g e of the searches and since these c a r d s u s u a l l y had the most entries their use w a s o f t e n the s l o w e s t . A s a result, u s u a l l y o n l y one or, at the most, t w o searches c o u l d be c o n d u c t e d s i m u l t a n - eously a t the i n d e x . T h e f a c t that the absence of one U n i t e r m c a r d f r o m the file n u l l i f i e d m u c h of the u s e f u l n e s s of the entire index also m a d e the prospect of l o s i n g a U n i t e r m c a r d f r i g h t e n i n g . A c a r e f u l analysis of the t w o h u n d r e d test reports w a s then u n d e r t a k e n to see w h y the difficulties w e r e experienced and if they c o u l d be c o r r e c t e d . T w e n t y - t h r e e reports or 1 1 . 5 % of the t o t a l w e r e a d e q u a t e l y in- d e x e d . A l l of these w e r e concerned w i t h concrete t h i n g s : design, f a b r i c a t i o n , testing, etc. N o n e d e a l t w i t h t h e o r e t i c a l subjects. T h i r t y - n i n e reports or 1 9 . 5 % of the t o t a l c o u l d be s a l v a g e d if the reports w e r e divided up into separately n u m b e r e d sections so as to avoid the m a n y false drops. T h e s e 3 9 reports had to be separated into 3 3 4 sec- tions. T h u s the 2 0 0 reports u n d e r s t u d y a c t u a l l y represented 495 u n i t s of w o r k , a d o u b l i n g in the e f f e c t i v e size of the r e p o r t c o l l e c t i o n to be i n d e x e d . T h i r t y - f i v e reports ( 1 7 . 5 % of the 2 0 0 ) w e r e m a d e r e t r i e v a b l e by e x e r c i s i n g m u c h t i g h t e r c o n t r o l of the U n i t e r m s by such means as e l i m i n a t i o n of s y n o n y m s , the addi- tion of cross references, definitions, quali- f y i n g phrases and t h a t w h o l e a p p a r a t u s of s u b j e c t h e a d i n g c o n t r o l w h i c h C o o r d i n a t e I n d e x i n g c l a i m s to e l i m i n a t e . D i f f i c u l t y w a s experienced w i t h 5 4 r e p o r t s ( 2 7 % ) because the c o o r d i n a t i o n of m u l t i p l e related U n i t e r m s p e r t i n e n t to the s u b j e c t m a t t e r of a d o c u m e n t c o u l d n o t be v a r i e d . T h a t is to say, if f o u r U n i t e r m s in a certain com- bination w e r e necessary f o r the r e t r i e v a l of a report, v a r i a t i o n s of these U n i t e r m s , t h a t is u s i n g t w o or three of the terms or c h a n g - i n g their o r d e r g a v e m e a n i n g s w h i c h bore 280 COLLEGE AND RESEARCH LIBRARIES no relationship to the m a t e r i a l s o u g h t . I n other w o r d s , the mere accretion of terms did n o t a l w a y s m a k e an idea m o r e specific but r a t h e r c h a n g e d m e a n i n g s . Y e t it w a s necessary to l e a v e the t e r m s " f r e e " since some of the w o r d combinations did fit the s u b j e c t m a t t e r . T h i s p r o b l e m tended to o c c u r w i t h h i g h l y technical s u b j e c t s and could be solved o n l y by p r o v i d i n g e l a b o r a t e definitions and e x p l a n a t i o n s l i m i t i n g the m e a n i n g s of the U n i t e r m s especially w h e n used in certain combinations. T h e s e w a r n - ings and guides w e r e f a r m o r e c o m p l i c a t e d t h a n the usual e x p l a n a t o r y statements f o u n d in c o n v e n t i o n a l s u b j e c t h e a d i n g lists f o r the same s u b j e c t m a t t e r . A l m o s t a l l the reports s h o w e d instances of m i s c o o r d i n a t i o n of a d j e c t i v e s , sub-head- ings and c o m p o u n d terms. F o r e x a m p l e , a p a r t i c u l a r c h e m i c a l report, i n v o l v i n g e l e v e n elements, required the use of three a d j e c t i v a l U n i t e r m s , t w o sub-heading-type terms, and one other t e r m , a l l of w h i c h could apply to all the elements. E a c h of these a c t u a l l y applied to o n l y one or t w o of the eleven ele- ments, b u t c o u l d be c o o r d i n a t e d w i t h a l l . Y e t it w a s not p r a c t i c a l to subdivide the report i n t o sections to p r e v e n t the m a n y false drops. T h i r t y - e i g h t reports ( 1 9 % ) had difficulties because a d j e c t i v e U n i t e r m s c o u l d be misapplied. Sub-headings, espe- c i a l l y c o m m o n ones like " d e s i g n , " " p r o p e r - ties," and " c a l c u l a t i o n s , " caused t r o u b l e w i t h 7 9 reports ( 3 9 - 5 % ) . C o m p o u n d s u b j e c t headings split i n t o U n i t e r m s led to false drops in 1 4 4 reports. ( 7 2 % ) . T h i s w h o l e question about the false drops, or as it is sometimes r e f e r r e d to, the " n o i s e " in this i n f o r m a t i o n scheme has been the sub- j e c t of some of the most heated debates a b o u t U n i t e r m i n d e x i n g . T h e proponents of C o o r d i n a t e I n d e x i n g a d m i t t h a t there are false drops b u t t h a t these occur i n f r e q u e n t l y and are so w i d e l y scattered and statistically insignificant that they can be n e g l e c t e d . O n the other hand, some theoretical studies of the p e r m u t a t i o n s and c o m b i n a t i o n s possible w i t h a g r o u p of U n i t e r m s h a v e s h o w n that e x t r e m e l y l a r g e n u m b e r s of false drops c o u l d be expected. I t has been a m u s i n g to note that the same m a t h e m a t i c s used by D o c u m e n t a t i o n I n c . to i l l u s t r a t e the diffi- culties c a t a l o g e r s experience in s e l e c t i n g the p r o p e r p e r m u t a t i o n s of m u l t i p l e - t e r m de- scriptions in an alphabetic index, is also used in the t h e o r e t i c a l studies to d e m o n s t r a t e the n u m b e r of false drops to be expected w i t h C o o r d i n a t e I n d e x i n g . 7 I t is obvious, of course, that the n u m b e r of false drops w h i c h m a y be experienced w i l l v a r y g r e a t l y f r o m sample to sample. F o r e x a m p l e , reports w h o s e s u b j e c t m a t t e r is w i d e l y scattered w i l l p r o d u c e v e r y f e w false drops, w h e r e a s reports w h i c h are con- centrated in e x t r e m e l y n a r r o w s u b j e c t fields w i l l p r o d u c e m a n y false drops. S i m i l a r l y , simple short d o c u m e n t s w h o s e i n f o r m a t i o n is confined to t w o or three concepts w h i c h can be described in a f e w w o r d s w i l l p r o d u c e v e r y l i t t l e " n o i s e " in an i n f o r m a t i o n system, but l o n g treatises c o v e r i n g m a n y v e r y tech- nical topics, w h i c h require m a n y U n i t e r m s to describe them a d e q u a t e l y , w i l l p r o d u c e a d e a f e n i n g r o a r . I t is obvious then that each collection of i n f o r m a t i o n w i l l cause its o w n c h a r a c t e r i s t i c n u m b e r of false drops if U n i - t e r m i n d e x i n g is used. T h e r e f o r e , at one of the A E C l a b o r a t o r i e s a g r o u p of f r e q u e n t l y used subjects w a s a n a l y z e d to see h o w m a n y false drops C o o r d i n a t e I n d e x i n g w o u l d p r o d u c e . Since u r a n i u m corrodes v e r y readily, the usual practice is to can it or clad it w i t h other metals. T h e r e is, as a result, an 7 M o r t i m e r T a u b e , C . D . G u l l and I r m a S . W a c h t e l . Unit Terms in Coordinate Indexing, T e c h n i c a l Report no. 3 . . . f o r the A r m e d S e r v i c e s T e c h n i c a l I n f o r m a - tion A g e n c y , N o v e m b e r , 1952; also as chapter 4, p. 37, T a u b e , op. cit. A s an unpublished appendix to his " E v o l u t i o n or I n v o l u t i o n ? " , M r . M o r r i s a n a l y z e d an artificial g r o u p of 36 reports all of which could be covered by 5 U n i t e r m s . S e t t i n g up all the possible intelligible combinations he could w i t h these 5 U n i t e r m s , M r . M o r r i s w a s able to " r e g e n e r a t e " 92 valid r e f e r e n c e s ( 3 9 % of the total) 81 f a l s e drops ( 3 4 % of the total) 27 " c o n f u s e d " concepts ( 1 1 % ) and 39 " f a r - f e t c h e d " concepts ( 1 6 % ) . JULY1955 281 e x t e n s i v e l i t e r a t u r e in the A E C libraries on the corrosion of u r a n i u m , the corrosion of c l a d d i n g metals, the c o m p a t i b i l i t y of u r a n - ium w i t h these metals, and so on. A r e p o r t on the corrosion resistance of a c l a d d i n g m e t a l , t h e r e f o r e , o f t e n contains a d d i t i o n a l i n f o r m a t i o n on v a r i o u s properties of u r a n - ium, b u t no i n f o r m a t i o n on the corrosion of the u r a n i u m . A t the time of the C o o r d i n a t e index test there w e r e 1 7 5 v a l i d r e f e r e n c e s in the c a t a l o g to the corrosion of u r a n i u m . H o w e v e r , a c o u n t of the corrosion r e f e r - ences to a series of m e t a l s p r o d u c e d the f o l l o w i n g n u m b e r of false leads to u r a n i u m the c a t a l o g and they w o u l d be h a r m f u l o n l y if there w a s an a c c u m u l a t i o n at any one point. N e v e r t h e l e s s , they g i v e some idea of the ratio of false drops to v a l i d leads one m i g h t e x p e c t . c o r r o s i o n : Number of cards in catalog No. of Uni- term false leads ' A N T I B O D I E S — Radiation Effects I I 6 E N Z Y M E S — Radiation E f f e c t s 5 1 7 3 M E T A B O L I S M — Radiation Effects 2 9 62 N U C L E I C A C I D S — Radiation Effects IO 26 I O I 167 A L U M I N U M — C O R R O S I O N (300) cards produced 102 false leads Z I R C O N I U M — C O R R O S I O N (200) cards produced 44 false leads S T A I N L E S S S T E E L — C O R R O S I O N (250) cards produced 70 false leads T o t a l false leads to U R A N I U M — C O R R O S I O N = 216 A c t u a l number of entries f o r U R A N I U M — C O R R O S I O N = 175 I n o t h e r w o r d s , three s u b j e c t areas w i t h a t o t a l of some 7 5 0 r e f e r e n c e s p r o d u c e d 2 1 6 false leads to U R A N I U M - C O R R O S I O N . A n d this did not e x h a u s t the possibilities, f o r the corrosion of all the other m e t a l s : copper, tin, z i n c , etc. w o u l d also h a v e pro- duced their q u o t a of false drops. T h u s , in this instance, the false drops w o u l d ex- ceed the v a l i d r e f e r e n c e s . O n the other h a n d , a check of a sub- h e a d i n g applied to a l p h a particles p r o d u c e d o n l y a l i t t l e o v e r 2 % of false leads to g a m m a r a d i a t i o n . C o n v e r s e l y , seven re- ports on the a l p h a emission of a m e r i c i u m 2 4 1 p r o d u c e d three false leads to t r i t i u m , three to n e p t u n i u m and one each to isotopes of cesium, u r a n i u m , c o b a l t , fluorine, nio- b i u m , a r g o n and to d e u t e r i u m . T e s t s w e r e also m a d e on a series of b i o l o g i c a l s u b j e c t h e a d i n g s u s i n g a reverse a p p r o a c h ; n a m e l y h o w m a n y false leads w o u l d sample h e a d i n g s p r o d u c e u s i n g the same s u b - h e a d i n g ? T h e s e false drops w o u l d , of course, be scattered t h r o u g h o u t E v e n a l l o w i n g f o r duplications, the n u m - ber of false leads thus p r o d u c e d w o u l d be i n t o l e r a b l e . I t is t r u e t h a t the s u b j e c t s chosen m a d e f o r a v e r y severe test of U n i t e r m s . T h e y w e r e not a v e r a g e samples. H o w e v e r , they w e r e n o t chosen because they w o u l d p r o d u c e the l a r g e s t n u m b e r of false drops. T h e y w e r e chosen because they are v i t a l to the w o r k of the A E C l a b o r a t o r i e s and it is in these s u b j e c t areas, and others like them, t h a t the libraries m u s t p r o d u c e the necessary i n f o r m a t i o n . I f C o o r d i n a t e I n d e x i n g can- not help here, then it has l i t t l e to o f f e r the A E C libraries. C o u l d these false drops be e l i m i n a t e d by v a r i o u s d e v i c e s ? I n p a r t , yes. T h e y can be m i t i g a t e d s o m e w h a t by s e g m e n t i n g c e r t a i n reports and n u m b e r i n g the sections separ- a t e l y . I n one test g r o u p , almost 2 0 % of the reports thus c o u l d be s a l v a g e d . T h i s , h o w - e v e r , m a t e r i a l l y increased the c a t a l o g i n g , f i l i n g and p o s t i n g w o r k loads. A n o t h e r s o l u t i o n w a s to use p o l y t e r m s . T h i s means 282 COLLEGE AND RESEARCH LIBRARIES essentially a b a n d o n i n g U n i t e r m s and u s i n g f u l l - s c a l e s u b j e c t headings w i t h none of the a d v a n t a g e s of the c a r d c a t a l o g . A g a i n this w o u l d have to be applied to those v e r y sub- j e c t areas w h i c h are used most f r e q u e n t l y and are of the g r e a t e s t i m p o r t a n c e , l e a v i n g U n i t e r m s f o r the less used m a t e r i a l s w h i c h can be c o n t r o l l e d a d e q u a t e l y by the simplest c a t a l o g i n g . In spite of a l l e f f o r t s to m o d i f y the i n d e x i n g in o r d e r to p r e v e n t false drops, 45 reports in a sample of 200 still p r o d u c e d false drops.8 I n a d d i t i o n to false drops, several of the tests w e r e t r o u b l e d w i t h the problem of " l o s t " i n f o r m a t i o n . I n other w o r d s , co- o r d i n a t i o n of U n i t e r m s f a i l e d to p r o d u c e i n f o r m a t i o n w h i c h w a s k n o w n to exist. I n one test 1 6 . 5 % of the reports proved ir- r e t r i e v a b l e . 9 T h e r e seemed to be three rea- sons f o r t h i s : 1. A l l the essential aspects of the report w e r e not covered by the title, section headings, and abstract. W h e r e the i n d e x i n g w a s , in accordance w i t h the U n i - term i n s t a l l a t i o n m a n u a l , confined to the title and abstract, in over 1 6 % of the cases the indexers had to g o back and c a r e f u l l y read the report and re-index it. 2. O c - casionally the v a r i o u s levels used d i f f e r e n t terms instead of c o o r d i n a t e d U n i t e r m s . T h i s w o u l d happen w i t h reports t h a t cov- ered i n d i v i d u a l components of a l a r g e r ap- paratus. Since the name of the l a r g e r ap- p a r a t u s did not occur in the title or abstract, a U n i t e r m f o r it w a s not p r o v i d e d . A g a i n , this w a s solved by a p p l y i n g s t a n d a r d cata- l o g i n g practices and s u p p l y i n g U n i t e r m s even t h o u g h they did not appear p r o m i n e n t l y in the r e p o r t . 1 0 3. T h e t h i r d reason w a s a 8 I n the B r i t i s h tests, R o y a l A i r c r a f t E s t a b l i s h m e n t , op. cit., t h e r e w a s , i n g e n e r a l , one f a l s e drop f o r each r e l e v a n t item f o u n d . 9 T h e B r i t i s h test, R o y a l A i r c r a f t E s t a b l i s h m e n t , op. cit., reported 1 5 % lost items. H o w e v e r , u s i n g their U . D . C . c a t a l o g , t h e y lost 5 0 % of the s e a r c h e d i t e m s ! 1 0 A s e x a m p l e s of this, t a k e n f r o m A S T I A c a r d s , one m u s t u s e both U n i t e r m s F I L A M E N T S a n d C A T H - O D E S to find all the filament r e f e r e n c e s ; one m u s t s e a r c h C R Y S T A L S , C R Y S T A L L O G R A P H Y , Q U A R T Z a n d m a n y other h e a d i n g s to find the neces- s a r y c r y s t a l r e f e r e n c e s . O f c o u r s e s u c h difficulties can be o v e r c o m e by p r o v i d i n g c r o s s - r e f e r e n c e s , but it does psychological one. P e o p l e are accustomed to certain p a t t e r n s f o r s u b j e c t headings and they do not t h i n k to look u n d e r such s t r a n g e terms as H I G H , H A L F , U L T R A , 90, B , R U B B I N G , V E R Y , H , 1 9 5 2 and so on. T h e r e is also a f o u r t h r e a s o n : some people do not f o r m u l a t e their questions v e r y pre- cisely, b u t close in on their desired references by t r y i n g o u t v a r i o u s s u b j e c t headings. T h i s f o r m of b r o w s i n g and r e l y i n g on suggestions is p r a c t i c a l in an o r d i n a r y c a t a l o g w h e r e titles and abstracts are a v a i l a b l e , but cannot be done in a U n i t e r m c a t a l o g . D i r e c t i o n of action b e t w e e n coordinated U n i t e r m s also caused difficulties. In one test sample, 1 2 % of the reports indexed demonstrated this d i f f i c u l t y . T h e s e w e r e p r i m a r i l y chemical reports i n v o l v i n g possible reversible chemical reactions and physics re- ports i n c l u d i n g p a r t i c l e reactions. T h e solu- tion f o r this w a s to use p o l y t e r m s or " b o u n d " terms, w h i c h is j u s t a n o t h e r w a y of s a y i n g o r d i n a r y s u b j e c t headings. I n addition, some m i n o r problems w h i c h c o u l d lead to difficulties s h o w e d up in the tests. T h e visual c o o r d i n a t i o n of l o n g c o l u m n s of n u m b e r s w a s f a t i g u i n g and caused m a n y errors. Some means of m a - chine coordination is necessary f o r extensive searches. M o r e m e c h a n i z a t i o n m u s t also be developed to speed up posting. U n l i k e f i l i n g mistakes, p o s t i n g e r r o r s w e r e e x t r e m e l y d i f - f i c u l t to detect and caused t r o u b l e . 1 1 U s e r s o b j e c t e d to the lack of selectivity. A person w a n t i n g a g e n e r a l paper on an entire field is also b u r d e n e d w i t h references to all the specific papers on p a r t i c u l a r aspects of the field. T h e o n l y w a y to c u l l this mass is to take the e x t r a step of c h e c k i n g the accession cards. i l l u s t r a t e the h i g h d e g r e e of s c a t t e r i n g that the U n i t e r m s i n d u c e . 1 1 I t m a y be n e c e s s a r y to adopt some v e r i f i c a t i o n sys- tem a s u s e d i n b a n k posting or in p u n c h c a r d w o r k . T h i s w o u l d i n v o l v e e i t h e r c h e c k i n g a carbon tape or d u p l i c a t e posting. I n either case it w o u l d m e a n a d o u b l i n g of posting time. JULY, 1955 283 O n the positive side of the l e d g e r , C o - o r d i n a t e I n d e x i n g did save o n e - t h i r d of the c a t a l o g space. T h i s w o u l d m e a n the re- p l a c e m e n t of a 9 6 d r a w e r 3 x 5 card file by nine o r so 5 x 8 d r a w e r s . T h e a c t u a l in- d e x i n g w e n t v e r y f a s t , p r o v i d e d the i n d e x e r did not g o beyond the title, section headings and a b s t r a c t . T h i s , h o w e v e r , o f t e n led to inadequate i n d e x i n g . I n a c t u a l practice, m o r e time w o u l d h a v e to be spent by the i n d e x e r r e a d i n g the report. S t i l l , it is be- lieved the i n d e x i n g w o u l d be faster t h a n routine c a t a l o g i n g . I t c o u l d not be determined if C o o r d i n a t e I n d e x i n g could help retrieve m o r e i n f o r m a - A Reply to T h e r e are a great many points in D r . W a r - heit's paper to which w e might take excep- tion, but there are only f o u r specific matters which urgently require clarification. 1. A l t h o u g h the initial statement of D r . W a r h e i t ' s position refers to a "series of stud- ies" of both a theoretical and practical nature and although D r . W a r h e i t indicates that his paper is based to a large extent on actual tests rather than on theoretical considerations, most of the figures in the paper are based on a single test in which only 200 items w e r e in- dexed and posted. T h e r e is presented at the end of these remarks the tabulations made from that test, and it can be noted how many of D r . W a r h e i t ' s figures are taken from this tabulation. 2. T h e original L o s A l a m o s test concluded that U n i t e r m indexing w a s inadequate index- ing leading to misconception, partial retrieval, false drops, etc. T h i s conclusion is reiterated by D r . W a r h e i t . In describing the conditions of the L o s A l a m o s test, those w h o conducted the test stated: In this i n i t i a l study, T a u b e ' s r u l e s w e r e f o l l o w e d as c l o s e l y a s p o s s i b l e : U n i t e r m s w e r e a s s i g n e d f r o m the titles, a b s t r a c t s , contents lists, a n d o c c a s i o n a l l y p a r a g r a p h h e a d i n g s . T h e r e p o r t s w e r e not r e a d . T h e p r o b l e m s of s y n o n y m s a n d h o m o n y m s w e r e c o n s i d e r e d of little i m p o r t a n c e and w e r e g e n e r a l l y d i s r e g a r d e d . tion by its m o r e flexible specificity and t h r o u g h its increased n u m b e r of access points. T h i s w a s due to the f a c t that the l i b r a r i a n s w e r e w o r k i n g w i t h fixed samples w h e r e the k n o w n , against w h i c h the un- k n o w n w a s being m e a s u r e d , w a s derived f r o m the s t a n d a r d c a t a l o g . B e c a u s e of the v e r y n e g a t i v e results of these p r e l i m i n a r y e x p e r i m e n t s , the A E C li- brarians decided to suspend a n y f u r t h e r w o r k on U n i t e r m I n d e x i n g and to a w a i t the results of field trials n o w b e i n g m a d e at the C o l l e g e of A e r o n a u t i c s , C r a n f i e l d , E n g - l a n d , and w i t h A S T I A c a r d s in this coun- t r y . Dr. Warheit N o references are given for these amazing statements. N o such rules or anything like them can be found in anything w e have written, nor have w e been able to trace the source of this mis- conception. In D r . W a r h e i t ' s paper there is a reference to the Installation Manual (no page citation) as the source of the instruction to index reports without going beyond the title and abstract. W e have gone over the Installation Manual, w o r d by w o r d , trying to pin down the source. A l t h o u g h there is an explanation of the method of setting up a U n i t e r m Index based on standard cards sup- plied by other indexing organizations (the title and abstract on a card can be used if they give more information than the subject- headings), it is difficult to believe that D r . W a r h e i t and the L o s A l a m o s people could have confused the indexing of a document with the conversion of a completed index to U n i - term form. Y e t , this is the most charitable ex- planation w e can devise for an error which is as gross as it is gratuitous. I t is certainly not necessary to conduct tests to arrive at the conclusion that sloppy and partial indexing makes bad indexes. 3. N o valid conclusion about posting can be drawn from handling 200 reports which re- quired the creation of 910 U n i t e r m cards. A s the vocabulary stabilizes (and neither D r . W a r h e i t nor the L o s A l a m o s testers deny that this w i l l happen) the rate of posting even under the w o r s t conditions w i l l increase by 284 COLLEGE AND RESEARCH LIBRARIES several hundred per cent. T h e r e really w a s no need for D r . W a r h e i t to base his conclu- sions on such a small sample. T h e people in charge of information at A E C headquarters in W a s h i n g t o n have inspected an installation in which close to 50,000 items have been posted. T h e y know that cards can be pre- sorted and that postings can be accumulated on punched cards. Even with a simple manual system, different posters can post different parts of the alphabet; and a w e l l tabbed manual system has enabled us to average be- tween t w o and three postings per minute, i.e., 120 to 180 per hour as contrasted w i t h D r . W a r h e i t ' s figure of 40. 4. I t is a recognized limitation of the U n i - term System or any machine system of coordi- nate indexing that it does not readily permit browsing in a subject arrangement of titles and abstracts. D r . W a r h e i t ' s remarks on this point may serve to bring it to the attention of those w h o missed it in our report on The Evaluation of Information Systems. B u t here again it seems that his comments cannot be based on any actual tests because A E C se- curity regulations do not permit readers to browse through the catalog. T h e contribu- tion which the U n i t e r m System makes to improved security would counsel its adoption by the A E C above all other agencies, even if it w e r e only barely as effective as the present system. Hence, I feel that it is unfortunate that the A E C librarians see nothing good in the system. A c t u a l l y , the only "controversy now raging about this novel form of index- ing" and up to now the only completely un- favorable published articles on the Uniterm System are products of A E C l i b r a r i a n s . — Mortimer Taube, Documentation, Inc. T a b u l a t i o n o f an E v a l u a t i o n o f C o o r d i n a t e I n d e x i n g for 200 L A R e p o r t s Adequately indexed Required modification: N e e d for report division N e e d for b e t t e r U n i t e r m selection P r o b l e m o f scientific terminology N e e d for p o l y terms A d j e c t i v a l S u b - h e a d i n g - t y p e C o m p l e x headings P r o b l e m o f direction Adequately index after modification Still presenting problems of specificity after modification Irretrievable after modification * T o t a l n u m b e r of r e p o r t s indexed = 200 T o t a l n u m b e r of u n i t e r m s used = 910 T o t a l n u m b e r of t i m e s u n i t e r m s w e r e used = 2 2 8 3 Selected Reference Books of 1954-55 (Continued from page 2J7) R e p o r t s % o f total U n i t e r m s R e p o r t s * I n v o l v e d % o f total U n i t e r m s * 2 3 I I . 5 — — 177 88.5 — — 3 9 1 9 - 5 — — 3 5 1 7 - 5 84 3 . 6 8 5 4 2 7 . 0 108 4 - 7 4 — — 1 3 4 4 58.82 3 8 1 9 . 0 164 7 . 1 7 7 9 3 9 - 5 184 8 . 0 5 144 7 2 . 0 996 4 3 . 6 0 2 4 1 2 . 0 4 4 i - 9 3 9 9 4 9 - 5 — — 4 5 2 2 . 5 — — 3 3 1 6 . 5 — — the 1940 edition ( G u i d e V 3 5 6 ) of this criti- cal and selective bibliography to include ma- t e r i a l — b o o k s and a r t i c l e s — w r i t t e n since 1940 concerning the humanities and the social sci- ences in Japan. T h e same classified arrange- ment has been used, but an entirely new sec- tion on W o r l d W a r I I and the Occupation, 1941-52, has been added. F u l l bibliographical details are given f o r each item, and brief descriptive notes f o r some. T h e r e is an index to titles and personal names. F u l l titles of abbreviated periodicals, are listed under " P e r i o d i c a l s " in Chapter I I I . — M . C . C o r d i e r , H e n r i . B i b l i o t h e c a sinica . Author Index to the Bibliotheca sinica of Henri Cordier. ( 2 d . edition, 4 vols. P a r i s , 1904-08. S u p p l e m e n t , 1 v o l . , P a r i s , 1 9 2 4 ) . C o m p i l e d , issued and dis- t r i b u t e d by the E a s t A s i a t i c L i b r a r y , C o l u m b i a U n i v e r s i t y L i b r a r i e s . N . Y . , 1 9 5 3 . 84P. $ 3 . 2 5 . F o r C o r d i e r see Guide V 1 9 1 . T h i s w o r k provides a much needed index to this valuable bibliography. JULY, 1955 285