To the homepage of Prof. Dr. Walther
Umstätter
To the publicationlist 1970 - 1985 - / -
To the publicationlist 1986 - 1990
To the publicationlist 1991 - 1995 - / -
To the publicationlist 1996 - 2000
Published in: Proceedings Fachbereich BID der FH Hannover 17.-19. Oktober S.398-408 (1989)
There are two typical alternative illusions in the field of librarianship:
1. Lancaster and some others are believing in a paperless world in the near future.
2. Old fashioned librarians seem to believe that catalogue cards can not disappear.
In reality we observe that from day to day more and more information is only available on an international host, a CD-ROM, an inhouse system in a local area network, or on a private micro-computer. Notwithstanding, the production and usage of paper is growing steadily. So we have to realise that information resource management in the next decades has to handle all these different types of information storage systems in an optimised proportion. That means to know the advantages and disadvantages of printed materials, of microforms, of magnetic and optical storage systems, as well as the problems of information transmission.
Also, questions about the usefulness of electronic mail, fax machines or character recognition readers are important topics of modern teaching. The economic aspect of information with all the time consuming searches, the often ineffective visits in different libraries and the attempts to be better informed than the competitors on the market or in the scientific community, makes the old fashioned libraries obsolet and requires a modern knowledge based system. The important question in the near future is not, how students can learn to sort catalogue cards in the one correct way, but how we get a deeper and wider insight into the possibilities of finding the available European information sources. It will be not so necessary to think about the second or third keyword for a book, but to have experience in free- and full-text searching. Distributed computer intelligence, handling of graphics, and the estimation of information demands will dominate the conceivable evolution.
In my opinion it is time to think more about the reduction of noise and redundancy to get real information, than about the development of printed card oriented cataloguing rules from the last century. The European market, the scientists and the common people need directly available low-priced information (fiction or non-fiction) not only from books or journals, but also from knowledge bases or expert systems. Information resource management has to consider that creativity, serendipity and associative thinking should be supported by sufficient active and passive offers at the right place. Centralised systems for document supply like BLDSC, TIB, or Mitterrand's inaugurated TGB "Tres grande Bibliotheqe" seems to be adequate for conventional sources. The aim of European programs like Adonis, Apollo or Artemis make other developments information logistics more feasible. One of the most unknown and influential elements in this divergent perspectives will be the control of copyrights.
What we have to do here, in my opinion, is only to look into the near future. It's clear, nobody knows anything precisely about the next decade, in which Europe shall consolidate and modernise his structures. But what we can do, is to look into the history to find out the most important trends, which will determine the developments in libraries, documentation centres and in the archives. Undoubtedly, information technology has brought a dramatic change into our world. This is a crucial trend, but what we would like to know is the impact of all these gadgets concerning our profession.
A look back to take a glimpse of the future:
The strongest trend in the last centuries was the doubling time of printed items, which was estimated with roughly 16 years by Fremont Rider(1) a half century ago. This explosion of publication, often called "information explosion" is nearly unchanged till now, and there is no indication that it can be reduced significantly without information technology. So the only thing that has retarded the growth of regular book and journal production in the last years is the transition to other forms, like self-printed reports and author's editions just as new storage media like microfilm, microfiche, and magnetic or optical disks for example.
The often cited Delphy study from 1980 published by Lancaster et al. with the title : "The Impact of a Paperless Society on the Research Library of the Future." made the assumption that in the year 1995 50% of the technical reports will only be available in electronic form(2). Another point, mentioned by Lancaster, is the projection of AT&T, that there will be a billion terminals in North America by the year 2000(3). Already a fraction of this value would be enough to change the scientific community.
Notwithstanding I agree with Donald King who has written: "Lancaster's view of the future has many desirable features, although I am not convinced that a completely paperless system is as inevitable - or desirable - as he asserts."(4) To take Lancaster's own words: "there is no real question that completely paperless systems will emerge in science and in other fields."(5). In consequence of the exponential growth, there is no chance to hope that a library will be able, over a long period, to grow with the same speed as the world production of papers and books in science, fiction, and non-fiction. There is not enough money to buy all the sources, not enough space to store all the volumes, and not enough time to make them available. We know that consumer price index from 1967 to 1987 increased by the factor 3.4 and the average cost per journal (in the Brandon list) by a factor of 6(6).The increase in average price per medical book rose by 439% from 1965 to 1989(7).
Even the Library of Congress has a more and more reduced percentage of total world wide available information. So we have to see, that the growing number of repositories at the end of the nineteenth century was followed by the creation of documentation. Card catalogues should help to discover remote relevant books and to make them available by ordering, and central catalogues became installed to establish interlibrary loan (ILL). For example, figures from OCLC might lead one to think that there had been an even greater increase in interlending in the USA, but this must be measured against the increasing number of libraries joining the network(8).
At the NLM (National Library of Medicine) 63% of all requests are received via DOCLINE, but from 1984 to 1987 NLM's fill rate for serials requests has declined from 78% to 67% (9).
A third synergistic factor which promotes this development is electronic document delivery. The financial break-even point at this time appears to be approximately 10,000 fax-transactions per year and per node for Dollar 100,000 at the MIT(10). In the Federal Republic of Germany we observed in the last year a doubled number of fax-machines from 150,000 to more then 300,000 today. That is fewer than ten percent of the world wide use.
Within the last 5 years Canadian libraries have increasingly abandoned telex in favour of electronic mail when sending ILL requests(11). Such mailbox systems will play also an important role at the electronic campus. As an example, the Carnegie Mellon University displays in an electronic bulletin board called LIBRARY the current table of contents for more than 33 scholarly journals(12). In a European academic network it will be possible to communicate with such mailboxes. The presupposition for such expensive interlending activities is a better knowledge about the source that has to be ordered. At the beginning of the 20th century the Dewey Decimal system, the Universal Decimal Classification, and the Library of Congress system were founded for a better content analysis. The libraries have changed their function, from an open shelf system to a hidden storage entirely. Lower and lower was the proportion of open accessible volumes and the amount of books that had to be ordered by using a catalogue was growing steadily.
1. Statement: The function of libraries is more and more identical with the function of documentation centres.
Documentation in the first half of our century was characterised by a very low level and can not be compared with that after the leap in 1963, what I would like to call the step into modern documentation. At this time the third computer generation was exclaimed by IBM, and the Weinberg Report(13) recommended to promote data bases like Chemical Abstracts, ERIC, Library of Congress, MEDLINE, NTIS or Science Citation Index by the government. It should be remembered, that in this report librarians were pictured as being somewhat anachronistic, as not keeping up with modern developments in automation or in new methods of bibliographic control. One indignant librarian at that time ceremonially burned a copy(14).
All these well known online available data bases have grown out of computer-based publishing, in which the material was keyboarded for computer typesetting. The rough estimation of a terabyte, that means ten with the power of thirteen bits information in the Library of Congress, at that time led to the realistic view, that it would be better to index all this materials and not to try to store it in full-text. Today we know, that the information content was underestimated more than thousand times, because Weinberg and his colleagues did not consider the information content of pictures and graphics. With more and more interest in the graphic capabilities of modern computer technology we learn to understand to eliminate the redundancy e.g. by image analysis and pattern recognition. Mass storage systems like hard disks with hundreds of megabytes, WORM and erasable disks give us the possibility to offer in special cases full texts or abstracts.
2. Statement: Gigabytes and terabytes of free storage space makes an electronic library with computer terminals like convenient entrances in the next years much more realistic than classical librarians can imagine.
Personal costs shows a doubling time of nearly seven years in industrialised countries since a long time, in contrast to the half-life of thirty months for prices that have to be paid in information technology. That means that in 1995 the relation will be ten times better for the technological handling of information than today and in the year 2002 over one hundred times. A very important point in this connection is the fact, that scanning machines will be very helpful to put thousands of pages completely into retrieval systems.
A short download from "The Electronic Encyclopedia", the CD-ROM of 1988 can be cited here. In this article Jessica L. Harris(15) pointed out:
"In the future, a researcher who is engaged in experimental work and needs the advice of colleagues in other locations may go to a computer terminal and send a message concerning the work in progress. The colleagues, at their own terminals, would receive the message, and reply at their convenience. This dialogue could continue for some time, until the researcher's work was completed, whereupon a report would be prepared at the same terminal and transmitted by computer to the editor of a journal, to be considered for publication. After the editor receives it, he or she would transmit it to reviewers who would judge its publication worthiness and send their comments back to the editor, who would synthesise them for the author. After revising the paper, the author would retransmit it; the editor would then arrange for it to go through composition and publication in the journal. At the same time the paper would go to the abstracting and indexing services that covered it in their publications. A scenario such as this one is technically possible at present; all these steps can be carried out at electronic speeds on the basis of only one keyboarding. Omission of the final step of paper publication is also possible, with substitution of direct transmission, as in SDI, to workers who have registered an interest in the topic of the paper."
The electronic library at the Bell laboratories is using some of these features in the last years, and thousands of scientists in the world exchange their disks with each other and also with their publishers. Networks like EARN, JANET or DFN are also heavily used in this electronic world. It's a danger that librarians are not involved good enough in this landscape today.
Let us suppose an output of 10 gigabyte of important readable characters per day in Europe, so it's not a matter of storage technology to collect all this materials. It is at first a matter of organisation and at second a matter of data arrangement. Both of these problems have to be seen in regard of copyright and of intellectual property. Most of the important new texts are created in a digitalized form. There would be no problem to make them available in international networks or on CD-ROM if there is a real demand and if the publishers can be sure to earn enough money.
3. Statement: A great part of older undigitalized information can be imported into machine-readable form by optical character readers (OCR).
There are different offers today of texts like the Bible, Goethe, Shakespeare, the philosophical works of Kant or encyclopedias and many others. From day to day the number is growing. The advantage of such electronic books is so significant, that everybody who has only little experience in this field will be convinced that there is a great demand. In connection with Hypertext, Hypercard or Hypermedia the benefits for instance on CD-ROM, CD-I or DVI are immense. Full-text retrieval requires not only software enhancements like hedges, highlighting, a search within n sentences etc.(16), it is also necessary to partition the books, periodicals, or reports into intellectual units that can be handled without greater problems. The hypertext philosophy will be helpful in this text preparation and automatic or half-automatic indexing will be inevitable.
4. Statement: One of the most important tasks for information resource managers will be to handle electronical stored full-texts.
All these different possibilities to store, to offer and to reorganise information make it necessary to think about an European information resource management.
Let us look at a rather simple problem:
A normal person who would like to monitor all information in her special topic of interest, out of the journal literature, is getting perhaps twenty most important papers from her own three subscriptions per year. Following the experience she knows that the regular screening of some thirty journals in her special library brings additional forty papers of her interest. After that she tries to visit five other libraries to browse sporadic fifty named periodicals to gain in total eighty publications. With this sources she extrapolate Bradford's law of scattering and estimates the total amount with roughly two hundred papers. So she knows that sixty percent of her demand has to be retrieved by citation analysis or by other search methods after the screening of eighty journals. As we know, a physicist is reading 190 Articles per year as an average(17).
In our example, more than thousand pages have to be ordered by interlibrary loan from local libraries or a document supply centre (DSC). Such a DSC is the well known British Library DSC from which Scandinavian libraries receive 50% of their international loans. If system interface were improved, the existing European DOCLINE link between Chalmers University Library in Gothenburg, TIB (Hanover) and BLDSC in Boston Spa could be extended in scope(18). The role of the "Bibliotheque de France", that shall be built till 1995 for far more than a half billion ECU, can be expected.
To facilitate co-operation, more than 100 libraries in North America use CONSPECTUS, an online subject inventory of library collections(19). In Europe CONSPECTUS is tested since 1984 by the British library(20). In this context it should be remembered that in 1980 ADONIS was largely stimulated by the desire to solve the copyright problem. At that time 10,000 commercial publishers in Europe were active. This whole industry exceeded revenues of 20 billion ECU (most probable 30 billion today). Parallel to ADONIS, the initial concept of ARTEMIS was to test the existing network technology for full-text transmission. As a Result, the following APOLLO project investigated the feasibility of a satellite-based system for document delivery(21). There are some different experiments in electronic publishing and document delivery in Europe(22). A taxonomy of copymarks is already proposed by Michael Spring from University of Pittsburgh(23). The development in electronic production, exchange and perception of papers will be determined by the European copyright law and by the question wether it can be proved.
5. Statement: Electronic document delivery for whole Europe needs broad band communication like satellites and for short distances optical fibres. On the legal side it must be justified.
Graphics, pictures and tables has to be stored and transmitted in the same way as characters. The massive increase in the use of collections is leading to a physical deterioration of many volumes, so that some libraries are already having to restrict the use of some endangered titles(24). Also the Library of Congress stored some of the mostly used materials on optical discs for testing this new technology and for protection of the printed material(25). But we know that high resolution document images need over 1000 times more electronic storage space than byte coded characters. Reduction of redundancy and image compression has to be understood on principle, because it will be possible to have a retrieval not only for words and phrases but also for colours, patterns or outlines. The indexing of graphics and pictures on the basis of pattern recognition is possible. Such a colour image database for an ethnology museum was build up for 30 billion Yen in Japan in the last ten years(26). We can also find additional experiences in the field of botanical and zoological cladistics.
Notwithstanding there are some doubts that optical disks are the instant solution to the problem of mass document storage that is frequently assumed(27). Rather it will be necessary to distribute editions with a high demand on many European strategic points in form of printed volumes, microfiche, or optical discs. These system of document supply has to be optimised. Following the statement of Donald W. King(17), that the current journal system is inefficient, because ten copies of articles are distributed for each use, we should be aware, that ADONIS will show, that the number of photocopies per article and library is not so immense as some publishers want to make believe. With other words, it is nonsense to buy and to store a volume with more than twenty papers, if only one is used by patrons.
6. Statement: User's demand has to be investigated more precisely. So we have to find out which information in which form has to be offered at which place and at which time economically.
A Scientist can save as an average 8% of his time and money if he neglects reading and the browsing of journals, books and bibliographies. It may be a result of bad experiences with libraries and interlending that some of them are doing so, but such scientists are in the great danger to be misinformed and to produce rubbish. In Britain Ensor estimated, that industry wastes nearly 10 million ECU a year on old information(28). This value seems to be much too low, because Gränzer(29) estimated that in the Federal Republic of Germany nearly 8 billion ECU will be lost by wrong information logistics. That means, that the information was not at the right time on the right place.
In many cases we find the cause for this lack of information not only in the too low level of acquisition in our libraries, moreover it's a lack of knowledge about the existing sources. So we know from the statistics that German university libraries have an acquisition comparable with that of the USA, but we can find only one half. Another problem is the distribution of information in an industrial plant, a rural area, or a university. The Integrated Academic Information Management System (IAIMS) started in 1984 and partially supported by NLM grant in 6 universities, is a basic approach to this problem. For example the IAIMS workstation at John Hopkins is the prototype of the so called Knowledge Workstation which helps to move easily between clinically oriented and literature databases to retrieve facts, refresh memory, compose a book chapter, develop and maintain a knowledge base, and communicate with colleagues(30). We should be aware, that in such an environment also new operating systems like those of transputers, with distributed intelligence, OCCAM is only one of them, will take place. This is also of relevance for efficient text scanning(31).
In the same direction of fifth-generation parallel-processing computation evolves the Electronic Information Delivery Online System (EIDOS) from OCLC. EIDOS is using photocomposition tapes received from publishers as sources of machine-readable books. It was found out that 48% of book use is of 15% of a book or less(32).
7. Statement: What we need today is a knowledge base at each information point that can tell us the different types of sources, their costs, their availability and their use.
A short statement of the question of standardisation should be done here. The MARC format as well as the German MAB2 format was created a quarter of a century before. It is with all his varieties very common and there are made some attempts in retrospective conversions of card catalogues(33) or scanned full-texts(34). A proposal for an extension of MARC to a modernised METAMARC exists already(35). But in my opinion Europe should try to get a real modern format, that is able to take over old MARC or MAB2 formats and which is modern enough to face the next decades with full-text, expert systems, and electronic publishing. The different adaptations to ISO/OSI in many European countries, the USA or Canada seems at the moment also very confusing.
8. Statement: International library standards has to be newly created or, like the MARC format, to be modernised.
After all I think there is no doubt, that electronic handling of information is much more important than library education till now makes clear. Understanding of artificial intelligence, system analysis, operations research, programming, and database construction is necessary in Harold Borkos opinion. He emphasized that these specialized education should be made in arrangement with such faculties and shouldn't be a part of the library school curriculum(36). Some believe also that librarians should become teachers in the use of information technology(37). That's possible but in my opinion we should concentrate at first to our primary problems. We should find out the real users demand, working out a plan for information logistics and use information technology only to fulfil our duty. That some basic knowledge in information technology is necessary is obviously. There are some European examples for an Information resource management curriculum in Glasgow, Maastricht and Sheffield, but this is not enough. In the "New Directions in Library and Information Science Education" Jose-Marie Griffiths and Donald W. King emphasized: "It is clear that we must prepare for an information-dominated future. Essential components of this future are librarians and information professionals who will be severely challanged by the Information Age."(38).
(1) Rider, F.: The Scholar and the Future of the Research
Library.
Hadham Press, New York (1944)
(2) Lancaster, F.W.; Drasgow, L. and Marks, E.: The
Impact
of a Paperless Society on the Research Library of the Future.
Final Report to the National Science Foundation. Urbana, Il.
University of Illinois, Graduate School of Library Science (1980)
(3) Lancaster, F.W.: The Future of the Library in the Age
of Telecommunications.
Telecommunications and Libraries: A Primer for Librarians and Information
Managers.
White Plains, NY: Knowledge Industry Publications, Inc. (1981)
(4) King, D.W.: Roadblocks to Future Ideal Information
Transfer
Systems.
Telecommunications and Libraries: A Primer for Librarians and Information
Managers.
White Plains, NY: Knowledge Industry Publications, Inc. (1981)
(5) Lancaster, F.W.: Whither Libraries? or Wither
Libraries.
Coll. & Res. Lib. 50 (4) 406-419 (1989)
(6) Kronenfeld, M.R. and Gable, S.H.: Update on inflation
of journal prices: medical journals, U.S. journals, and Brandon/Hill list
journals.
Bull. Med. Libr. Assoc. 77 (1) 61-64 (1989)
(7) Brandon, A.N.; Hill, D.R.; Levy, G.L. and Levy, J.W.:
Selected list of books and journals for the small medical library.
Bull. Med. Libr. Assoc. 77 (2) (1989)
(8) Cornish, G.P.: Interlending and document supply. A
review
of recent literature.
XIV. Interl. and Doc. Supply 16 (3) 103-109 (1988)
(9) Lacroix, E.M.: Impact of DOCLINE on interlibrary loan
service at the National Library of Medicine.
Bull. Med. Libr. Assoc. 77 (1) 42-47 (1989)
(10) Reintjes, J.F.: Application of Modern Technologies
to Interlibrary Resource-Sharing Networks.
JASIS 35 (1) 45-52 (1984)
(11) Lunau, C.D.: Canadian advances in the application
of
electronic mail and interlibrary loan automation.
Interl. and Doc. Supply 16 (2) 58-60 (1988)
(12) Tinsley, G.L.: An electronic bulletin board:
LIBRARY.
Spec. Libr. 80 (3) 188-192 (1989)
(13) Weinberg-Kommission:
Science,
Government and Information.
Report of The President's Science Advisory Commitee USA. Washington
(1963)
(14) Weinberg, A.M.: Science, government, and
information:
1988 perspective.
Bull. Med. Libr. Assoc. 77 (1) 1-7 (1989)
(15) The Electronic Encyclopedia", CD-ROM version (1988)
(16) Basch, R.: The seven deadly sins of full-text
searching.
Database 12 (4) 15-23 (1989)
(17) King, D. W.; D. D. McDonald, N. K. Roderer :
Scientific
journals in the United States. Their production, use, and economics.
Hutchinson Roß Publ. Comp. Stroudsburg (1981)
(18) Winkel, A.: The application of new technology to
interlending and document supply in Scandinavia - a progress report.
Interl. and Doc. Supply 16 (3) 101-102 (1988)
(19) Ferguson, A.W.; Grant, J. and Rutstein, J.S.: The
RLG
Conspectus: Its Use and Benefits.
Coll. & Res. Lib. 49 197-206 (1988)
(20) van Heijst, J.: Conspectus und die nationale und
internationale Bibliothekspolitik.
in: Nationalbibliotheken im Jahr 2000, S.81-95 Hrsg.:v. Köckritz, S.
und Nowak, K., Buchhändler- Vereinigung GmbH, Frankfurt am Main
(1988)
(21) Gurnsey, J.: Electronic Document Delivery - III
Electronic
Publishing. Trends in the United States and Europe.
Learned Information, Oxford (1982)
(22) Mastroddi, F.A.: Experiments in electronic
publishing
and document delivery: Results of the EEC's DOCDEL programme.
Interl. and Doc. Supply 16 (4) 121-128 (1988)
(23) Spring, M.: The origin and use of copymarks in
electronic
publishing.
J. of Doc. 45 (2) 110-123 (1989)
(24) Wheelhouse, H.: Resource sharing - A critical view
of the literature.
Interl. and Doc. Supply 16 (4) 136-143 (1988)
(25) Videodisc and optical digital disk technologies and
their applications in libraries. A Report to the Council on Library
Resources,
Inc. Wash. by Information Systems Consultants Inc.
ERIC Reports ED 257 433 (1985)
(26) Hong, J.-K.; Hashihara, H.; Ioka, M.; Kurokawa, M.;
Sato, M. Sugita, S.; Kubo, M. and Yamamoto, Y.: A Color Image Database
for
an Ethnology Museum.
Cologne Computer Conference Sept. 7th-10th, 1988 Volume of Abstracts IBM
B.5-1 - B.5-2 (1988)
(27) Moralee, D.: Facing the limitations of electronic
document
handling.
The Electronic Library 3 (3) 210-217 (1985)
(28) Blagden, J.: Do we really need libraries?
Saur, Bingley (1980)
(29) Gränzer, W.: Was uns bewegt.
IBM Nachrichten 37 S.18 (1987)
(30) Lucier, R.E.; Matheson, N.; Butter, K.A. and
Reynolds,
R.E.: The Knowledge Workstation: An Electronic Environment for Knowledge
Management.
Bull. Med. Libr. Assoc. 76 (3) 248-255 (1988)
(31) Cringean, J. K.; Manson, G.A.; Willet, P. and
Wilson,
G.A.: Efficiency of text scanning in bibliographic databases using
microprocessor-based, multiprocessor networks.
J. of Inf. Sci. 14
335-345 (1988)
(32) Kilgour, F.G.: An essential information delivery
system.
in: ASIS '87. Proceedings of the 50th Annual Meeting of the American
Society
for Information Science. Boston, Mass. Oct. 4-8, 1987 Vol. 24 134-137
(1987)
(33) Harrison, M.: Retrospective conversion of card
catalogues
into full MARC format using sophisticated computer-controlled visual
imaging
techniques.
Program 19 (3) 213-230 (1985)
(34) Hein, M.: Optical scanning for retrospective
conversion
of information.
The Electronic Library 4 (6) 328-331 (1986)
(35) Hinnebusch, M.: METAMARC: An Extension of the MARC
Format.
Inf. Tech. and Lib. 8 (1) 20-33 (1989)
(36) Borko, H.: Artificial intelligence and expert
systems
research and their possible impact on information science education.
Educ. for Inf. 3 103-114 (1985)
(37) Moore, M.: Innovation and education: unlimited
potential
for the teaching library.
Bull. Med. Libr. Assoc. 77 (1) 26-32 (1989)
(38) Griffiths, J.-M. and King, D.W.: New Directions in
Library and Information Science Education.
Greenwood Press, Inc. American Society for Information Science (1986)