Cover Image

Not for Sale



View/Hide Left Panel

The Efficiency of Metallurgical Abstracts

NERIO GAUDENZI

To discuss the efficiency of metallurgical abstracts, after what was said by Frank T.Sisco at the 1957 Annual Meeting of the National Academy of Sciences, may appear as an absurdity, or at least as presumptuous. In accordance with the statements of the Director of the Engineering Foundation of New York, it probably would be advisable to speak about a failure of metallurgical documentation, and of documentation in general in the abstracting field, if, as Mr. Sisco said, “If a research job in the USA costs less than $100,000, it is cheaper to do it than to find out if it has been done before and is reported in the literature.” (5).

But while we wait for results, which may be available in 1960, of five-year studies, promoted by the ASM in 1955, in the pilot plant at the School of Library Science at Western Reserve University, to determine if machine methods of searching and finding coded metallurgical literature are practical, it will be not useless to point out the actual situation.

It may be that in America automation, when introduced in the metallurgical literature searching field, will produce revolutionary results, or at least results more satisfactory than those achieved up to now with usual methods. In Europe, however, the conditions are quite different, and probably it still will be prudent for many years to try to improve the defects of existing documentation systems.

Our statements, which are the fruit of direct researches published previously, (4), are the result of work carried out by one who is both a producer and a user of abstracts in the field of metallurgy. They confirm or support the candid but substantially exact evaluation of Mr. Sisco.

We must recognize first that technical documentation has had in these last years a considerable development and that, therefore, there exists at the present time abundant material for studies which, unfortunately, being spread in many streamlets, are still very little known and are not adequately used. Standards and prescriptions resulting from careful studies in the field of bibliographical

NERIO GAUDENZI Editor-in-Chief of the journal Alluminio, ISML, Milan, Italy.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 393
--> The Efficiency of Metallurgical Abstracts NERIO GAUDENZI To discuss the efficiency of metallurgical abstracts, after what was said by Frank T.Sisco at the 1957 Annual Meeting of the National Academy of Sciences, may appear as an absurdity, or at least as presumptuous. In accordance with the statements of the Director of the Engineering Foundation of New York, it probably would be advisable to speak about a failure of metallurgical documentation, and of documentation in general in the abstracting field, if, as Mr. Sisco said, “If a research job in the USA costs less than $100,000, it is cheaper to do it than to find out if it has been done before and is reported in the literature.” (5). But while we wait for results, which may be available in 1960, of five-year studies, promoted by the ASM in 1955, in the pilot plant at the School of Library Science at Western Reserve University, to determine if machine methods of searching and finding coded metallurgical literature are practical, it will be not useless to point out the actual situation. It may be that in America automation, when introduced in the metallurgical literature searching field, will produce revolutionary results, or at least results more satisfactory than those achieved up to now with usual methods. In Europe, however, the conditions are quite different, and probably it still will be prudent for many years to try to improve the defects of existing documentation systems. Our statements, which are the fruit of direct researches published previously, (4), are the result of work carried out by one who is both a producer and a user of abstracts in the field of metallurgy. They confirm or support the candid but substantially exact evaluation of Mr. Sisco. We must recognize first that technical documentation has had in these last years a considerable development and that, therefore, there exists at the present time abundant material for studies which, unfortunately, being spread in many streamlets, are still very little known and are not adequately used. Standards and prescriptions resulting from careful studies in the field of bibliographical NERIO GAUDENZI Editor-in-Chief of the journal Alluminio, ISML, Milan, Italy.

OCR for page 393
--> abstracting are very often ignored by those who should apply them. Recommendations formulated during international meetings and by qualified associations, and prescriptions elaborated by standards institutions often remain a dead letter, and old lines already outmoded by more recent requirements continue to be followed. This imbalance is due also to the lack of a bridge between the researcher and the documentalist. The bridge theoretically is built by the technical documentalist, who in Italy usually is a graduate in engineering, in chemistry, or in the physical or mathematical sciences and who, owing to casual circumstances, is obliged to deal with documentation problems; or he may be a graduate of the faculty of arts who, always for professional reasons, has to deal with technical documentation, which is looked upon as a part of general documentation. The first, coming from the technical-scientific faculty, is well acquainted with disciplines he is graduated in, but is not sufficiently a master of the documentation techniques, since the knowledge acquired in the university in this field is absolutely inadequate. The second, coming from faculties of arts (philosophy and languages), might well have the mental structure and fundamental characteristics of a bibliographer but, having no specific knowledge of the subject, cannot have adequate competence to resolve the concrete problems of documentation as applied to physics, chemistry, engineering, etc. Let us consider now the documentation and especially abstracting in the field of metallurgy. This consideration could serve as an example of combining the requirements of the documentalist with those of the researcher in the field of metallurgy. The problem of the efficiency of bibliographical abstracts in metallurgy, as well as in any other branch of science, can be examined from a double point of view, i.e., quantitatively and qualitatively. Quantitative efficiency Apparently quantitative efficiency is not difficult to evaluate. Theorically, it is sufficient to know the number of journals of metallurgy published in a year and the number of articles contained in them. If examination of bibliographical abstract journals shows that all articles published have been abstracted, we can say that the network of abstracts is perfect, since not a single article has escaped abstracting. The more the system deviates from this theoretical and ideal condition the lower is the efficiency of the service. Let us see what the situation is in practice. First of all, it is necessary to decide what we mean by a journal of metallurgy. We can use the definition adopted in our previous study (4, p. 15), which naturally is subject to criticism, on the basis of which metallurgical journals were considered to be “those journals

OCR for page 393
--> whose titles contain a clear reference to mining and metallurgy treated as a science and including everything from ore to finished products.” On the basis of long years of study on this subject, we believe that it is possible to calculate that there are, in the entire world, approximately 2000 metallurgical journals actually “alive.” This number can be taken as a starting point, but it is not complete; many articles of a metallurgical character are being published in journals of chemistry, physics, engineering, etc. For example, results from our earlier study (4, p. 32) show that in the field of light metals, in 1954, of the articles abstracted in various abstracting journals, 34% were reported from nonmetallurgical journals. This scattering of publications on a certain subject (in our case metallurgy) throughout an enormous number of journals ought to be reduced by authors to a necessary minimum. Naturally, in the case of metallurgical publications it is impossible to avoid the dispersion as, for example, an article on the use of aluminum in the textile industry belongs in a textile journal rather than in a metallurgical magazine, but articles on the science of metals or the technology of metallurgy should be published in journals of metallurgy. Also prompt notice of original articles published could be given by authors or by editors of journals to a central bibliographical association, charged with the collection of such data. In any case, based on the actual situation, if it is assumed that for various abstracting services all 2000 metallurgical journals will be examined, about a third of the articles published would escape and the efficiency would be only 70%. If we next consider the number of bibliographical abstract journals concerned with metallurgy, we can distinguish two types of publications: (a) those dealing exclusively with metallurgy, e.g., Metallurgical Abstracts of the Institute of Metals, or Extraits of the Revue de Métallurgie and (b) those having a general technical character and containing sections devoted to metallurgy, e.g., Chemical Abstracts, Chemisches Zentralblatt, Bulletin Signalétique du Centre National de la Recherche Scientifique. The first group of publications number about 70, if we judge from our first study (4, Appendix 1), based on the examination of publications existing on this subject and, above all, on the direct control of journals. In the second group, in which we naturally include only abstracting journals of wide coverage and completeness, can be mentioned the following: (1) British Abstracts. BI Chemical Engineering and Industrial Inorganic; (2) Chemical Abstracts; (3) Engineering Index (in English); (4) Chemisches Zentralblatt (in German); (5) Bulletin Signalétique du Centre National de la Recherche Scientifique (in French); (6) Boletin (del) Centro de Documentacion Cientifica y Tecnica, Mexico (in Spanish); (7) Referativnyi Zhurnal Metallurgia (in Russian). The total number of abstracting journals dealing with metallurgy can be taken as about 80.

OCR for page 393
--> If we now wish to establish a relation between original metallurgical journals and those publishing bibliographical abstracts of metallurgy, we find that this is about 2000:80, i.e., one abstract journal to every 25 original journals. This ratio seems normal because, according to available data contained in the literature, there should exist actually about 50,000 original technical and scientific1 and about 1400 abstracting journals (3), with a ratio of 35. An outsider might think that such a high number of abstracting journals in metallurgy might mean a higher efficiency since there would be a better guarantee that all articles published would be covered. Anyone who has had experience in this field, however, knows that the efficiency is quite limited because every journal operates independently; consequently, many articles are abstracted several times, while others are missed altogether. For example, in the field of light metals we learned in our previous study (4, p. 30) that some journals (Modern Metals, Chicago; Light Metals, London; Aluminium, Düsseldorf; Light Metal Age, Chicago; Alluminio, Milan) are covered by 10 to 12 abstracting services; the total almost certainly is at least 20, since our previous investigation was limited to specialized abstracting journals and the 5 primary publications mentioned above are regularly abstracted also by abstracting journals having a general character. In our opinion a satisfactory solution of the quantitative problem could be put on the following basis. Limit the number of abstracting journals to principal fields offering a specific interest for certain groups of industry or institutions, as it appears for example, from the following: (1) one of general character (science of metals), including also the field of mining and metallurgy; (2) one dedicated to iron metallurgy; (3) one to nonferrous metals; (4) one to light metals; (5) one to foundry technology; (6) one to welding; (7) one to corrosion and surface treatments; and (8) one to economics and statistics of metallurgy. Naturally, there should be as close coordination as possible to minimize duplication. Actually, the picture of metallurgical abstracting journals is as follows (4, p. 27): 1. Science of metals 17 2. Iron metallurgy 7 3. Nonferrous metals 9 4. Light metals 14 5. Foundry 3 6. Welding 6 7. Corrosion and surface treatments 8 Total 64 1   The Academy of Sciences of the USSR (6) examines about 12,000 journals, of which about 9700 are printed in Latin alphabet types, about 1900 in languages having Cyrillic alphabets, and the rest (about 400) in languages with other types. Among the group with Latin types, only about 300 are in metallurgy.

OCR for page 393
--> It is obvious that a considerable reduction in the number of these publications should be possible. Every journal should carry abstracts in 5 principal languages: English, French, German, Russian, and Spanish. Naturally, where there is a qualified international institution for intercommunication, a unique journal could contain abstracts in all 5 languages or in some of them (as is done by the International Welding Institute or the Bibliographic Bulletin for Welding and Allied Processes which is published in French and English). In other cases, a unique journal in a determinate language could deal very well with all eight subjects, as is done today in Russia where a centralized organization permits such an enterprise. The Referativnyi Zhurnal, which standardizes the abstracting service for all sciences and is divided into separate sections for each of them, has achieved a considerable degree of completeness. Its metallurgical section published 14,957 abstracts in 1956, whereas the most advanced service in the free world (ASM Review of Metallurgical Abstracts) issued only 8000. A most laudable enterprise has been announced this year by Acta Metallurgica—the publication in English of abstracts appearing in the part of Referativnyi Zhurnal, Metallurgy which covers original journals that are very difficult to find in western countries. The number of bibliographical journals of metallurgy could be reduced to 5, if based on a linguistic principle, and to 8 if based on the principle of contents, to reach the maximum of 40. At least 40 of the existing journals could then be eliminated, and the service, with minor dissipation of forces, could be improved. To improve the efficiency of metallurgical abstracts it will be necessary, from our point of view, to centralize the service under a single institution in order to avoid a duplication of work and to assure coverage of areas now neglected. In this connection, attention could be called to an interesting idea actually under study by CIDA (International Centre for the Development of Aluminum) with headquarters in Paris, of which the following countries are members: Austria, France, Germany, Great Britain, Italy, Sweden, and Switzerland. This organization has in mind to centralize the abstracting service in the light metals field, and we hope this will be realized soon and concretely. International cooperation in the field of bibliographical abstracting is indispensable, particularly in Europe, where, since funds for this kind of work are scarce, they must not be wasted. On the other hand, the less well-off countries must have access to information on the research carried out in richer countries, so that for the former documentation assumes a greater importance than for the others.

OCR for page 393
--> Qualitative efficiency Let us now examine the qualitative efficiency of bibliographical abstracts. As is known, there are today two types of bibliographical abstracts—indicative and informative. According to the definition of Unesco (2), “an indicative abstract is a short abstract written with the intention to enable the reader to decide whether he should read the original publication or article”; that is, this kind of abstract indicates only what an article is about without attempting to be a substitute for it. The searcher, if the subject is of interest to him, must himself procure the original publication. The informative abstract “summarizes the principal arguments and gives the principal data contained in the original publication.” In this case the abstract may substitute for the original article and the searching of the latter can be avoided. This distinction, theorically quite precise, actually is a complicated one in practice. For one thing, the compilers of the abstracts do not always have these definitions clearly in mind. Also, compilation of informative abstracts requires a special competence in the material and a knowledge of preceding literature, which only a specialized person can have. For example, in effect, only one who deals with and lives in the foundry atmosphere is able to compile really informative abstracts on this part of metallurgy. This is what happens to the compiler of abstracts, while the users and documentalists, who are not always specialists, will find it difficult to judge at once whether an abstract is an indicative or an informative one, except in limited cases. No documentalist, for example, will be in doubt about the type of abstracts (informative) published in the Extraits of the Revue de Métallurgie, which each take one or more pages of the journal, or about the indicative ones of the Metals Review of the ASM, which consist of only a few lines. In this case, the criterion of method of compiling abstracts may be linked with length: an informative abstract is normally extensive and an indicative abstract is short. Thus in practice it happens too often that length is used as the basis of discrimination between indicative and informative abstracts, this being a principle that is much easier and more comfortable to apply than one based on the method of compilation. We have made these points to prove that it is not easy to establish which type of abstracts is practically preferred in the field of metallurgy. From the direct examination of 56 abstracting journals (4, Appendix 1), using evaluations devised by us which do not always coincide with those of other bibliographical researchers, we concluded that 39 of the journals have

OCR for page 393
--> adopted indicative abstracting, 20 journals informative, and 31 journals semi-informative.1 This last group includes abstracts of medium size, compiled according to a criterion between the informative and the indicative, constituting a neutral zone where assignment as to the one or the other principal type of abstracts would be highly objective. The statistical principle does not furnish sufficient information, and this lack is not eliminated through taking the language as a basis. Of French journals, 19 carry indicative, 7 semi-informative, and 6 informative abstracts. In journals printed in the English language the principle of semi-informative abstracts prevails in 14, indicative in 12, and informative in 10. In the German language 8 journals are semi-informative, 4 are indicative, and 3 are informative. In journals in Italian and other languages all three types of abstracts have the same number—one. If, however, we confine our interest to abstracting journals of more common use and of greater diffusion, known by a great many researchers and documentalists, we will perhaps find more reliable information. In France, for example, we can clearly recognize one tendency, as the Revue de Métallurgie, official publication of the Société Française de Métallurgie, the Circulaires d’Information Techniques of the Centre de Documentation Sidérurgique, Fonderie, and the Documentation Métallurgique, are issuing quite long abstracts which are decidedly informative in character. The use of indicative abstracts is limited to the Bulletin Signalétique of the Centre National de la Recherche Scientifique. In the United States, on the other hand, the most important metallurgical abstracting service, that of the American Society for Metals, which publishes its abstracts in the monthly Metals Review and then republishes them in annual volumes of ASM Review of Metal Literature, has adopted the indicative form of abstracting. Furthermore, if we were able to find 10 informative abstracting journals, they would not be too extensive and would be closer to the semi-informative than to the strictly informative type. The abstracts of NACE which, although they are published both in the monthly journal Corrosion and in the Annual Bibliographic Survey of Corrosion, also tend to the indicative form of abstracting. If we further consider the Metallurgical Abstracts of the Institute of Metals, London (also published monthly in the Journal of the Institute of Metals) and the Abstracts of Current Literature (published in the Journal of the Iron and Steel Institute, London) we find that both of these have a semi-informative character, tending, as to length, more toward the indicative type of abstracting than to the 1   The sum of 39+21+31 is superior to the number of journals examined (56) for many journals have adopted more than one type of abstract.

OCR for page 393
--> informative. Thus, it is obvious that in English-speaking countries (United States and Great Britain) preference is being given to an indicative or, at least, to a semi-informative type of abstracting in the metals field. In Germany, a country in which bibliographical services are particularly well developed and appreciated, there is no bibliographical journal of metallurgy of general character, and one is obliged to consult the classical Chemisches Zentralblatt, corresponding to Chemical Abstracts in the United States. In the field of iron metallurgy there is the Zentralblatt der Hütten-und Walzwerke, included in the journal Stahl und Eisen, which is decidedly indicative in character; also numerous other journals devoted to particular fields of metallurgy have, in general, a semi-informative character, tending, however, more to the indicative than to the informative form. If we depart from the difference between indicative and informative abstracts, it is evident that in English-speaking countries there exists a better and more complete bibliographical organization as compared with that in France or in German-speaking countries. We now shall pass on to the examination of other elements which could be advantageous for establishing the value of abstracting services and of the abstracts themselves. In order not to digress too much from the main subject, we shall consider only the principal aspects that have general bibliographical value and interest, including particularly abbreviations of the titles of the abstracted journals, the coding systems adopted by various services, and the manner of presentation of abstracts (pages of journals or cards). Abbreviation of the title of the journal examined Where, because of space limitations, it is not convenient, in the abstracts, to reproduce the full title of the original publication containing the article abstracted, it will be advisable to use the following: DIN 1502, Germany, 1955 (1st edition 1940). Recommendation ISO R 4 of international character, May 1954 (inspired by the ISA Bulletin 23 of 1940). Code for Abbreviations of Titles of Periodicals, of 1953, Great Britain. AFNOR 2–44–002 of 1944, in France. In Belgium, India, Holland, Denmark, Spain, and Switzerland there are similar standards. In the field of metallurgy, judged by the results of an investigation carried out by us, no one of the services has shown any interest in following these standards, and abbreviations are still made in a completely arbitrary way. For

OCR for page 393
--> example, speaking only of the more important services,2 we find that one and the same title is abbreviated in many different ways, as: ASM Review of Metal Literature: Metal Treatment and Drop Forging. Journal of the Iron and Steel Institute: Met. Treatment. Light Metals Bulletin: Metal Treatment & Drop Forging. Metallurgical Abstracts: Metal Treatment. Stahl und Eisen: Metal Treatm. Drop Forg. If, instead, we consider abstracting journals of general character, having sections on metallurgy, we find for the same title the following abbreviations: Chemisches Zentralblatt: Metal Treatment Drop Forging. Chemical Abstracts: Metal Treatment and Drop Forging. Bulletin Signalétique du Centre National de la Recherche Scientifique: Metal Treatment. Referativnyi Zhurnal: Metal Treatm. and Drop Forging. It is difficult to determine why there has been no standardization of abbreviations in the field of metallurgy. It may be that the compilers of various services (especially the smaller and more specialized ones) simply ignore the matter or consider it not worth the trouble. It may also be true in the case of more important services, which began their publications before there was any standardization, that for practical reasons they do not want to change the system already adopted by them; also it may be that the standards themselves are not as easy to apply here as with ISO R 4. This document prescribes (paragraph 4), “The normal method of abbreviation consists in suppressing the last letters of the word, at least two, substituting them with a full stop.” As for example the word “Journal” can be abbreviated as Journ., Jour., Jou., Jo., J.; the possibility of five different solutions in one standard seems to permit excessive freedom of interpretation, contrary to the very spirit itself of standardization. Coding system The coding problem is highly difficult and complex in the field of bibliography and a system which can satisfy everybody probably can never be found. In this review, we would like to consider only what principles have been adopted in practice for bibliographical abstracts in the field of metallurgy. Above all, it is necessary to distinguish between the real classification of general validity and 2   We have adopted the principle of Varossieau (1) who indicates as important those bibliographical services which publish at least 4000 abstracts in a year.

OCR for page 393
--> a generic classification. To the first group belong, for example, the decimal system and the ASM-SLA. In this case every article receives a fairly long symbol, formed by letters or by numbers, which identifies individual articles and affords the searcher the possibility of forming a classification which represents a permanent bibliographical consultation source, in which every subject will be kept constantly up to date. Under these circumstances, no importance is to be attributed to the fact that the classification has a universal value (at least in theory), i.e., whether it can be used for all sciences and by all systems, or whether, on the contrary, its value is limited to a specific field like the ASM-SLA system for metallurgy. The main thing is that it will be possible with the classification to maintain constantly up-to-date files in which every single article will be easily traceable. There are other less subtle coding systems which we can call generic. These are limited to the constitution of comparatively wide subdivisions, in which all articles on related subjects find their places, as in our field, welding, corrosion, and finishing. This principle, as we shall see, is generally applied. The Metallurgical Abstracts of the Institute of Metals, to indicate one of the most important and authoritative abstracting services, follows this system. Articles are subdivided as follows: (1) properties of metals; (2) properties of alloys; (3) structure; (4) dental metallurgy; (5) powder metallurgy; (6) corrosion and related phenomena; (7) protection; (8) electrodeposition; (9) electrometallurgy and electrochemistry; (10) refining; (11) analysis; (12) laboratory apparatus, instruments, etc.; (13) physical and mechanical testing, inspection, and radiology; (14) temperature measurement and control; (15) foundry practice and appliances; (16) secondary metals, scrap, residues; (17) furnaces, fuels, and refractories; (18) heat treatment; (19) working; (20) cleaning and finishing; (21) joining; (22) industrial uses and applications; (23) history of metals; (24) metal economics; (25) health hazards; (26) miscellaneous; (27) bibliography; (28) book reviews. With this system, retroactive searching of abstracts naturally becomes increasingly laborious as the years pass. With this type of classification, the bibliographic research of a specific subject becomes very long, because it is necessary to examine all articles on related subjects (for example, for a journal of forging one would have to consult all items on working, year by year). Beyond the above, there are other classification systems—if they can be called such—which give only a numeration, generally progressive per annum, of articles abstracted: this has, above all, a practical value, as it can be used for a quick marking of abstracts by users of abstracts who are looking for an original article.

OCR for page 393
--> Based on these principles, and with data previously collected by us (4, Appendix I), we observe that among 53 metallurgical abstracting services: 33 have adopted a generic classification system 7 have adopted the decimal classification system 6 have no classification system 5 have their own classification system 2 have adopted the ASM-SLA classification system3 If the examination is limited to important services, we find that only one (ASM Review of Metal Literature) has adopted the ASM-SLA classification system, while all others content themselves with classifications of generic character. These data show that in the field of metallurgy, no one of the important services actually is using the decimal classification system. Among the less important services, 7 have adopted the decimal system. They are: 1. AGM Leichtbau (light metals), Germany 700 abst./year 2. Literaturbericht (light metals), Switzerland 500 abst./year 3. Archiv für Metallfinishing (surface treatments), Germany 300 abst./year 4. Hutnik (general metallurgy), Poland 100 abst./year 5. IMM Abstracts (mining & metallurgy), England 2800 abst./year 6. SMRE Abstracts of Current Publications (safety of mines), England 1000 abst./year 7. Zeitschrift für Metallkunde, Germany — Of these, some are not real bibliographical journals (1, 3, 4, and 7); No. 2 is more like a restricted bulletin, and only the 5th and 6th can be said to be important from the documentation point of view in the field of mining. Therefore, it can be stated that in effect the field of metallurgy (excluding mining) has practically completely abandoned the decimal classification system. Instead of the decimal system, the ASM-SLA is being used by ASM, which is its promoter, and by AIM (Italian Society of Metallurgy) for its Atti e Notizie AIM. Other special classification systems are being adopted, like that of the NACE in Corrosion, of the Istituto Internazionale della Saldatura in the Bibliographical Bulletin for Welding, that of the Centre de Documentation Sidérurgique in their Circulaires d’informations Techniques and Bulletin Analytique, and those of the Documentation Métallurgique and of the Svetsliteratur. 3   According to information we received from the AIM, French and German iron metallurgy associations also have decided to adopt the ASM-SLA classification system.

OCR for page 393
--> The majority of the services use instead more or less similar classification systems. Presentation of abstracts The presentation problem of abstracts, i.e., whether to publish abstracts on cards or on normal pages has considerable importance from the documentation point of view. The first type presumably should be preferred by the searcher of abstracts, as it permits maintenance of both author and subject files, which can be kept continuously up to date and consulted rapidly and easily. For this the user naturally must have a documentation office with specialized personnel—something not always possible for small societies. On the other hand, the card system is more difficult and more expensive for the producer of abstracts, who must have an extra service, the cost of which is almost never compensated for by requests of users. Our earlier study (4, Appendix I) showed that of 63 journals only 3 insert leaflets printed only on one side which can be detached and inserted on white cards to form a file. These are Aluminium, Berlin (about 150 abstracts/year), Documentation Métallurgique, Paris (500 abstracts/year), and Ciencia y Tecnica de la Soldadura, Madrid (500 abstr./year). Ordinarily, preference is given to presentation on regular pages of journals. At this point the punch card system must be remembered, which is assuming a considerable development in recent years, and has one of its concrete applications in the documentation of metallurgy. Since 1950, the NACE, Houston, Texas, has offered a regular subscription service of punch cards, containing abstracts of previously published articles; about 2000 cards are issued in a year, at a cost of $100. Also ASM is distributing punch cards that follow its own classification system. These are blank and the user must attach the abstracts. It must be noted that these two institutions parallel their card services with a continuing conventional service of abstracts presented in the normal form in their periodicals Corrosion and Metals Review, as well as in their respective annual publications. As to this use of punch cards, we believe that this is a very important improvement, though at the present time it means a certain expense for the user. The annual cost of NACE punch cards is about $100, while the corresponding volume, Bibliographic Surveys of Corrosion, costs $10 (members). The practical diffusion of this punch card system will tell us whether or not its advantages more than offset its higher cost; our present opinion is that they do.

OCR for page 393
--> Conclusions From these considerations, it appears clearly that the relatively low efficiency of metallurgical abstracts is due above all to the excessive number of abstracting publications which operate without any connection with each other, so that many original journals are examined by the same services, whereas other original journals escape the network of abstracts entirely. The international bibliographical standards are scarcely observed. With these limitations, it can be observed that English-speaking countries have better services, and that the most complete service belongs to Russia. The efficiency could be improved appreciably: (1) by centralizing abstracting services, which could be carried out, for the entire field of metallurgy or for single branches (light metals, welding, etc.), by institutions which already have adequate installations, with the cooperation within international orbit (initiative of CIDA), outside of government or bureaucratic interference; (2) by reducing the number of abstracting journals, which actually is excessive; (3) by studying efficient systems able to reduce the dispersion of articles of metallurgical character in journals belonging to other fields; and (4) by distributing abstracts on punch cards. REFERENCES 1. VAROSSIEAU, W.W. A survey of scientific abstracts and indexing services. Rev. Document 16, 25–48 (1949). 2.——, International Conference on Science Abstracting. 20–25 June 1949. Unesco, Paris, 1951. 3. BESTERMAN T. Index Bibliographicus, Directory of Current Periodical Abstracts and Bibliographies. Vol. I, Science and Technology, 3rd edition, Unesco, Paris, 1952. 4. GAUDENZI, N. I riassunti bibliografici nel campo metallurgico con speciale riguardo ai metalli leggeri: analisi critica. Quaderni monografici ISML no. 1. ISML, Milano, 1956. 5. SISCO, F.T. What’s in the literature. Metal Progress, 122–124, 1957. 6. Institute of Scientific Information of the USSR Academy of Sciences, List of Abbreviated and Full Titles of Technical and Scientific Publications. Academy of Sciences of the USSR, Moscow, 1957.

OCR for page 393
--> This page intentionally left blank.