Group Title: Miscellaneous publications
Title: The National Agricultural Library and you
Full Citation
Permanent Link:
 Material Information
Title: The National Agricultural Library and you a speech delivered at the 12th annual Library Workshop of the Florida Agricultural Experiment Station Library at the North Florida Experiment Station, Quincy, Florida
Series Title: Miscellaneous publications
Physical Description: 14 leaves : ; 28 cm.
Language: English
Creator: Lulich, Ljubo
Publisher: Florida Agricultural Experiment Station
Place of Publication: Gainesville, Fla.
Publication Date: December, 1964
Copyright Date: 1964
Subject: Information storage and retrieval systems -- Agriculture -- Florida   ( lcsh )
Genre: non-fiction   ( marcgt )
Statement of Responsibility: by Ljubo Lulich.
General Note: "December 1964."
 Record Information
Bibliographic ID: UF00090269
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 309841094

Full Text







A Speech Delivered at the 12th Annual Library Workshop of the
Florida Agricultural Experiment Station Library
at the
North Florida Experiment Station
Quincy, Florida


Ljub o

Lul i c h

Division of Indexing and Documentation
National Agricultural Library
U.S. Department of Agriculture



Miscellaneous Publications
New Series No. 2

Gainesville, Florida
December 1964





Lj ubo Lulich

ALTHOUGH I HAVE BEEN assigned to talk about the National Agricultural
Library in general and about the Bibliography of Agriculture in particu-
lar, still the topic of this workshop is the Primary Journal; so I want
to at least touch base. For indeed the primary journal is the main diet
of the Bibliography; in fact, were it not for this material, we in the
Bibliography would all be out of business.

We have just gone through a rather thorough and systematic survey
of USDA research scientists for the purpose of ascertaining their informa-
tion and library service requirements, so that we can better chart our im-
mediate and future course. This survey was a part of the total work of a
task force on library services, which was authorized by the Department
back in 1962. You may have heard about it. Briefly, it was called TASK
FORCE ABLE (ABLE acronym for Agricultural-Biological Literature Exploita-
tion) and was made up of research scientists, research administrators,
systems designers, and management and fiscal experts. The Task Force was
teamed into four groups: Systems Requirements; Systems Design; Cost; and
Report Writing. The work of the Systems Requirements Group is the more
interesting and immediately important because it was the group that con-
ducted the survey; so its findings will certainly carry an impact on the
current and future activities of the Library.

The user survey made by this group was designed especially for
USDA research scientists. This limitation on the category of intended
respondents was based on the fact that our primary responsibility is to
Department people, in Washington and in the field. Applying strict cri-
teria in determining the qualifications of "research" and "research sci-
entist", the group identified about 4,500 USDA people as being engaged
in research. That was in June 1962. This initial pool of research sci-
entists represented the entire spectrum of the Department's research
activities, meaning practically the whole range of science. Of these

4,500 scientists 1,900 were selected systematically to receive question-
naires. The rate of response was something to be envied by professional
pollsters and pulse-takers something like 85 percent thanks mainly
to the doggedness of the survey team but also to a little prodding by
agency administrators. Anyway, a draft report of the survey is now be-
ing put in final form. If all goes well, the report will be completed
by the end of summer 1964.

In making one or two observations on the survey, I will not be
revealing anything prematurely. As a matter of fact, many findings
merely confirm what has been suspected all along by those whose business
it is to keep score in the science information game.

The first elements to be identified are the principal sources
of information that research scientists have at their disposal. I will
only take a few of these sources and put them in each of two main cate-
gories. First those of a personal nature:

Conversations with colleagues
Attendance at meetings
Informal personal contact
Advice of superiors

Then those of a library nature:

Primary journals
Abstracting journals
Routing current literature
Reference books, textbooks, handbooks
Review articles
Library accession lists
Library card catalogs

Now each of these and other sources take on a degree of impor-
tance varying with the nature of information sought, e.g., keeping
abreast of published literature vs. retrospective searching. The rank-
ing of the various sources of information according to their importance
to research scientists will be treated in detail in the report. This
is not the place for a statistical monologue; besides, statistics should
be read, not heard. So I will limit myself to half a dozen elements
which seem to repeat themselves throughout the ranking scale for the in-
formation sources. The primary journal is almost invariably at the top.
In general the ranking of the elements is as follows:

1. Primary journal 6. Informal personal contact,
2. Conversations with including correspondence
colleagues 7. Routing of current
3. Attendance at meetings literature
4. Advice of superiors 8. Bibliographies
5. Abstracting journals

Sad to say, but not very surprising, library accession lists and lib-
rary card catalogs are way down the list.

I think this is enough to show once again that, for keeping
abreast of published literature and developments in one's field of in-
terest, there is no substitute for the primary journal.

I would summarize this portion with an observation which I
would formulate this way: Junior scientists read more, use the library
more, seek advice more, and attend meetings less than their senior coun-
terparts. Again, this is not at all surprising when you consider that
the juniors must work their way up, while the seniors are already there.

So much for still another "user study." Scientists have been
the object of some concentrated studies into their information-gathering
habits and behavior. Some pollsters have even resorted to clocking
their subjects, using stopwatches for the purpose. Others have inven-
toried the user surveys done in the past few years and come up with
several hundred of them, all but a handful absolutely worthless.

Meanwhile, what is the Library going to do about all this. Ev-
erybody nowadays is talking about computers as the panacea for all doc-
umentation ills. When the word documentation is used in the context of
research libraries, you can almost hear the humming of a whole string of
machines and visualize thousands of little lights flashing like crazy.
I would not be exaggerating much by saying that in almost any library
situation the words "computers" and "documentation" are used synonymous-
ly, in the same breath.

But what is "documentation" anyway.

An Indian librarian was once asked why they had so many documen-
talists and so few librarians, when in fact both were doing pretty much
the same kind of work. Well, he replied, a documentalist in India earns
7,000 rupees a year, while a librarian earns 5,000.

This is a very practical assessment of the situation prevailing
in the field of documentation. There is a great deal of honest and some
premeditated confusion about what we mean by documentation, information
processing, data processing, information storage and retrieval, science
communication, information communication, and these in addition to the
traditional library service. This is really not surprising when you
consider that we are being conditioned by push-button psychology, aided
and abetted in this, I must add, by the enthusiastic but not too dis-
interested computer salesmen. Some of us believe that instant know-
ledge, like frozen TV dinners, is just around the corner or that it is
already here. This is understandable, too, in view of the fantastic
things that computers and similar machines have been doing and promise
to do in the future. We have been hearing, for example, about automa-
tic abstracting and indexing; about machines translating one language
into another at unbelievable speeds, and idiomatically, too; about ma-
chines reading printed pages at hundreds of words per second; about
televising library card catalogs and pages from books and journals direct-

ly to the receiver on your desk; about printing entire libraries on a
dozen 3" x 5" cards; and other similar schemes. Some of these schemes
will perhaps come into library use in the not too distant future; such
as automatic abstracting and indexing; facsimile transmission of printed
matter; and optical scanners reading all kinds of text. But others are
likely to come about only after we have retired, and this is too far
ahead to worry about today.

For most of us it is the information about documents rather than
information itself that we must handle and disseminate. The success of
this work is measured by how effective we are in informing the research
scientist what we have that may be of interest to him, and telling him
enough about it to enable him to decide whether or not he wants to see
the real thing. The real thing is the document itself, and this is a
good place to define that as being any object on which information is re-
corded: book, journal article, technical report, unpublished manuscript,
picture, map, microcard, etc.

These objects have been piling up in libraries at an alarming
but not very surprising rate when you contemplate statistical determina-
tions that have been made by some experts in the field. It has been de-
termined, for example, that four out of every five scientists that ever
lived are alive today. These scientists, along with others who are dri-
ven to writing by all sorts of professional and economic motives, have
been producing literature in quantities calculated to double library
collections every ten years for the next several decades. But tuis is
the qu~:titative side of the coin and therefore less serious. The other
side the qualitative aspect is the more serious one. Libraries can
acquire tons of books and journals and stack them neatly on the shelves -
this is easy enough because only storage space is involved. But if these
materials are not first tagged well and intelligently so that we know
what's in them and can call for them when we need to, then the whole ef-
fort is lost. I don't think anyone will dispute the contention that
information is useless unless it is used.

It is this aspect of the so-called "information crisis" that
has been the main source of concern both to the research library admin-
istrator and to the research scientist, but for different reasons, of
course. The first cannot do the job properly because he doesn't have
adequate human and material resources, while the second is starving in
the midst of abundance, so to speak. The last metaphor is reflected in
the programs of scientific meetings which have tended with increasing
regularity to squeeze in some time to discuss the literature problem.

In any planning it is important to bear in mind that a research
library, as is also any vital system, is a composite of elements designed
to accomplish a purpose. Modifying one element invariably affects ano-
ther, with cumulative effects for the rest. The systems concept is not
peculiar to the manufacturing world; it applies to the library world as
well. Many libraries have tried to solve one element in their system
without regard for the rest, and have failed in the process. That is
why the Library administration in 1962 asked the Department to make a
feasibility study of our Library's activities, with a view to mechaniza-

Meanwhile, we have been conscious of the areas where some of
our documentation activities could be improved and expanded if only we
had the means. It does not require an inventory to reveal the short-
comings that are organic to the object on the face of it. Neither does
it mean, of course, that a critical look at the object by the audience
it is designed to serve will not reveal a different emphasis, a novel
direction, a modified orientation, or even questions about its useful-
ness. These are valuable products of evaluation and appraisal; in fact,
if they were not forthcoming, then the whole exercise would be a waste
of time and effort by everybody concerned.

One of the principal objects of the survey mentioned above was
the Bibliography of Agriculture. This is because the Bibliography is
virtually the sole visible link between a research scientist in the field
and us in the Library. I will describe the directions in which the Bib-
liography should go, as I see them, and leave the details about its cur-
rent status for later discussion. However, I do want to remind you that
the Bibliography of Agriculture is available free of charge to govern-
ment offices and public institutions. It is surprising that so many of
our own people don't know about this, even those who use it occasionally.
But this does not hurt nearly as much as the occasional revelation that
some of our people, who should know better, are absolutely unaware of
the existence of this publication.

The Bibliography will continue to be the main vehicle by which
the Library communicates with its national and international customers.
The printed volume is still the only instrument that can do the job ef-
fectively for most people, so until a better substitute is found I am
afraid we are stuck with it.

We are planning to increase the number of references indexed in
it from the current 100,000 level to an eventual 240,000 per year. We
hope to reach this goal by the late '60's. At that time the Bibliography
will be indexing an estimated 90 percent of the substantive agricultural
literature published in the world. This vast expansion in quantitative
coverage will have important implications for several activities of the
Library, but most of all for the format and structure of the Bibliogra-
phy itself. Of course we cannot be so dogmatic as to decree today what
its definitive form and content will be six years hence; we do not want
to be that rigid. We prefer instead to allow the evolutionary process to
have its way as much as possible. Nevertheless, we know the basic fea-
tures the Bibliography should have and the steps necessary to achieve

The Bibliography will continue to be a title-listing journal.
That is, we do not expect to be in the position, technically speaking,
to do abstracting ourselves. There is, however, an encouraging trend by
authors to supply abstracts to their papers, and by journal editors to
insist that they do. This sort of thing would certainly make us explore
the possibility of including authors' abstracts along with the titles.
Although some abstracting journals prefer abstracts done by specialists

other than the authors themselves, authors' abstracts are better than
no abstracts at all. There are also at least two other, more limited
alternatives meriting consideration. One is to choose a discipline,
such as Entomology, in which authors tend to supply abstracts to their
papers, and issue a separate abstract bulletin for that discipline.
Another is to choose a commodity such as has been done for wheat at the
University of Nebraska (Wheat Abstracts), and do the same for it. In
any of these and other alternatives the factors of economics are of
course decisive.

The Bibliography will continue to be a classified array of ref-
erences. This arrangement seems to suit most people. But classifica-
tion is troublesome, especially to the user who doesn't know all the con-
ventions and priorities employed by the classifiers. For example, you
might be a forester and so expect to find all references pertaining to
forestry in that section of the Bibliography. This is not so because
priorities put the material on tree physiology, diseases, etc., in the
Plant Science section; similarly, material on pests will be found in the
Entomology section. This inconvenience is usually minimized by supply-
ing a detailed subject index to the classified arrangement of references.
The Bibliography has only one such index which is issued in December.
To supply each monthly issue with its own subject index is therefore our
number-one priority. Let me assure you that setting this project as our
first priority is not an arbitrary decision on our part; we have been
hearing about this shortcoming for a number of years, from librarians
and from research scientists. Besides, the volume of references carried
in one monthly issue dictates that such a project be our prime concern.

A subject index in each monthly issue will enable you to locate
a small topic, or a minute aspect of a subject, quickly and reliably.
As things are now, you have to look over the titles in one or several
sections and guess the content of the article by its title alone, or you
must wait until the end of the year when the annual subject index is
issued. We plan to make this monthly subject index a cumulative feature:
on semi-annual, annual, and quinquennial basis, so that in July you
would be able to search the January-June issues in a single cumulation,
then the whole year in January, and finally a five-year cumulation in
one big volume.

The subject index will be examined carefully, from our viewpoint
as well as from that of our users, for possible division into several
parts. It might be useful, for example, to show all binomials in animal
and plant kingdoms (except economic plants and animals) in a separate in-
dex. These binomials scientific names of insects, microorganisms, non-
economic plants, etc. comprise about 60 percent of our entire subject
index, so we would be at least justified in thinking of them as warrant-
ing separate treatment.

We will also take a look at the geographic locations that are
now part of our index structure as a potential separate index for arti-
cles where the idea of geographic location may be of prime significance.
In these alternatives, we would have three subject indexes issued either
as separate sections within a single volume or as three separately pub-

lished volumes. There would be a strictly subject index showing pro-
ducts, concepts, techniques, applications, material, equipment, etc.;
a scientific name index for animal and plant kingdoms; and a geograph-
ic index. Each would cumulate semi-annually, annually, and every five
years. These cumulations would be done separately and independently of
other work, so that the regular publication of the monthly issues would
not be interrupted. This means that the publication program would be
a 12-month affair 12 regular issues of the Bibliography each year.
Today we must stop with the November issue because we have to turn every-
body to preparing the December index volume.

An essential part of subject indexing work is the authority
list or a volume showing the terms that are to be used for indexing,
how they are used, for what subjects and in what situations, and a host
of other instructions and directions. In most library situations this
volume is called the Subject Heading Authority List. Lately there has
emerged a version of the traditional thesaurus as the preferred form for
assigning subject terms. It is claimed that such a thesaurus is more
useful both to the indexer and to the user in leading them from one term
to another generically, specifically, by synonyms, by antonyms, and by
other relative guides.

One of our more glaring shortcomings has been the lack of any
such authority list or thesaurus. We do, however, achieve a degree of
uniformity in indexing by using the Bibliography's annual subject in-
dexes, plus numerous notes, decisions, and special instructions; and
throuFg periodic meetings of indexers. But these procedures do not give
us the kind of uniformity and consistency required by an authoritative
index. The need for an authority list or thesaurus is particularly ac-
centuated when it comes to using computers to prepare an index. We will,
therefore, start very shortly to compile a subject authority list or the-
saurus of terms used in indexing for the Bibliography of Agriculture.
Concurrently, we will look at the subject headings used on the cards in
our public card catalog and, hopefully, come up with a single, uniform
volume that will be the authority for subject analysis work for the whole
Library. This is not the case now: we use one set of terms for index-
ers and another set for catalogers, and the two sets are not entirely
compatible. I should mention at this point that the Library has recent-
ly issued a four-volume set showing the subject terms used for our card
catalog. This incidentally is the first time in the history of the Lib-
rary that such a compilation has been made.

The last item on the Bibliography that I will mention is its
subject coverage. In the planned expansion in this area the Bibliogra-
phy will go in at least three directions:

The first is to index more of the same. That is, we will index
the material we are receiving but are not indexing because we cannot
read the languages, or because the articles are of such form, content,
and treatment that we are not taking them at present, or because the
publication is older than a specified age, or simply because we cannot
fill the staff vacancies that occur.

The second is to recognize agriculture as a multidisciplinary

aggregate having more than concrete application to agriculture. In other
words, we are going to index more articles in microbiology, chemistry,
biochemistry, mycology, bacteriology, and similar disciplines. Today we
index articles in these fields only if they have specific agricultural
applications. The same applies to references in animal industry and
veterinary medicine dealing with research on diseases or conditions oc-
curring naturally in the animal, or research having veterinary implica-
tions. In doing this we are in effect saying that biological research
begins and stops with the animal object.

In agricultural products we do not take anything beyond primary
processing. This too is an abrupt place to stop, as agricultural econ-
omists would readily agree.

The third is to expand into fields that are now inadequately
covered: e.g., zoology; mammalian biology; hydrology; nutrition, includ-
ing experimental work, deficiencies, and diet therapy; family and home
economics; agricultural economics in the broad sense of the word; agri-
cultural meteorology; fishing and fish industry; and agricultural engi-

The expansion in subject and numerical coverage will bring the
Bibliography eventually to 800 or more pages per month. With this in
view, we will in good time explore the advisability of splitting up the
Bibliography into several logical groupings and publishing these groups
as separate journals. This seems to contradict the repeated allusions
to the interdisciplinary nature of science. It does indeed contradict,
so I ai using the word explore advisedly. Nevertheless, I think we
should see if we can come up with some such groupings as, for example:
Plant Science, Forestry, and Soil Science to comprise one journal; Agri-
cultural Economics, Agricultural Engineering, Agricultural Products, and
Animal Industry to comprise another; Entomology to make up a third; and
a fourth Miscellaneous group for the fields not fitting into the other

GOING NOW INTO OTHER areas of the Library's activities, the first in im-
portance are special bibliographies. A vital function of the Library is
putting the material it owns, and even the material it does not own, to
use. This it does in the form of bibliographies and lists on a single
subject, hence the name "special". While the Bibliography of Agriculture
merely announces what has been received in the Library, a special bibli-
ography sifts and sorts and often interprets, however uncritically, this

Special bibliographies are an important adjunct to any contem-
plated research; one that the Library is, or should be, admirably equip-
ped to provide. It doesn't make much sense for a research scientist to
spend his scarce and expensive time in literature searching for his re-
search project. This is neither efficient nor economical, and had better

be left to those who are equipped and trained to do just that. Unfor-
tunately, the Library has not been prolific in issuing these bibliogra-
phies. The reason is of course lack of staff. The situation has been
improving gradually, however, enabling us to issue recently a few special
bibliographies and lists that have been well received, such as Freeze-
Drying of Foods, Safflower Oil, and Japanese Serial Publications in the

Because this aspect of library function is so vital, the compil-
ation of special bibliographies and lists will be expanded proportionate-
ly more than any other activity. The larger staff to do the work that
only humans can do will be aided by machines to produce on demand any
legitimate bibliography or list efficiently and quickly. The system will
also produce, systematically and regularly, those special bibliographies
and lists that the Library will issue as a matter of policy.

ON THE SUBJECT OF translations, I must say that the Library's current
translation program is very limited in scope and in number of people
served. The majority of the Department's scientists must turn outside
for translation services or be satisfied with the cover-to-cover trans-
lations primarily of Russian language periodicals, provided they happen
to be available and pertinent to their work.

We do not attach undue importance to foreign research, but nei-
ther can we afford to be so conceited as to think that anything printed
in languages other than English cannot contain valuable information.
Since this information is not in English, and we are not exactly notor-
iously proficient when it comes to handling foreign languages, the in-
formation is not made available to those who should have it. I think the
situation is appreciated by all serious research scientists. Certainly
the inquiries received in the Library on the subject cannot be interpre-
ted other than as a mandate to institute a central translating service
for agricultural scientists in this country. We envision the Library as
headquarters for translation services and activities in the Department
and its field agencies, including also state experiment stations. But
we also appreciate the economics of such a program, knowing that trans-
lations are expensive, and therefore must be carefully controlled so
that no inadvertent duplication occurs, and only that which is necessary
gets translated. Only the affluent can afford indiscriminate transla-
tions, and Agriculture is not exactly famous for its affluence. We
would hope, therefore, at some time in the future when funds are avail-
able, to set up three options: Full Translations; Summary Translations;
and Individual Translation Service.

Full translations would be made of articles that are essential
in support of research.

Summary translation would be done for a research scientist who
may come across an article in a foreign language which he is not certain

is pertinent to his need. He could ask the Library for a summary of it.
We might write him a summary of it, or cut a record, or put it on tape,
or even telephone it to him. He could then decide whether he wants the
full translation or not.

Individual translation service would be similar to the summary
translation in that the translator and the requestor get together, with
the former reading the piece to the latter. If the requestor finds the
article significant a full translation is indicated.

ON THE SUBJECT OF a Printed Catalog, I might start by telling you that
early in 1963 we issued three samples of a bi-weekly publication called
Newly Cataloged Titles A Selected List. This was a listing of books
cataloged in the Library during the preceding two weeks. Quite a few
libraries issue such listings to keep their users informed about books
and journals that are being added to their collections.

In our case the experiment was a sort of trial balloon designed
to elicit reactions from about 400 research people. Circumstances forced
us to discontinue the sampling before we could gather a representative
response. Nevertheless, we view the trial sample as a very rough proto-
type of a publication that is on our active planning board. It amounts
to issuing a printed catalog in a series of publications that would show
what books and journals had been received during a specified period, and
how they were cataloged and classified. In addition to serving the pur-
pose of informing, the publication would also serve as a bibliographic
guide to catalogers and reference librarians in agricultural libraries.
Finally, these periodic publications would be cumulated into bigger vol-
umes, perhaps annually and again in five years, thus becoming a printed
record of books and journals added to the Library's collections during
that time.

One of the prerequisites of this plan is the closing out of our
present catalog, which has several million cards, and starting a new cat-
alog along with its parallel printed volumes. Another desirable feature
of the plan is the revision of our classification system for books. The
present system has outlived its logic and usefulness. This fact is appre-
ciated by most who have to use it, and especially by us who have to work
with it. But there is also general agreement that the conversion job to
a modern system is one of some magnitude and of course very costly. We
are acutely conscious of the need to do this eventually, and the sooner
the better for everybody.

THE VARIOUS ACTIVITIES I have mentioned could probably never be accom-
plished efficiently and economically without the use of modern machines,
so perhaps the subjects of mechanization and automation should have been

dealt with first instead of last, because they are the foundation with-
out which the planning would be largely an academic exercise. Realistic
administrators of research libraries, facing the tremendous increase in
the volume of literature to be controlled, coupled with the equally heavy
increase in the demand for special library service, have come to rea-
lize that the conventional library methods cannot do either job adequate-
ly. To keep adding more and more people, even if money were abundant,
is a partial and temporary answer, and is justified only for those work
routines that machines cannot do as well as humans, at least not at the
present stage of their engineering development. On the other hand, to
stand still and hope that the problem will somehow dissolve or go away
on its own accord is not only suicidal it is dishonest.

Like most service organizations, library service is like an
iceberg. I don't mean to say that it is cold and unfeeling; I mean
that most of what makes it work is submerged. It is a deceptive pic-
ture. Of course a library user does not look upon it exactly this way,
nor should he. He is only interested in the result, in how the library
meets his requests for information. How the library does this is not
one of his concerns. But let me tell you about this iceberg because
it symbolizes a series of routines and procedures that we will be ana-
lyzing to see which lend themselves to mechanization and for what pur-
pose. There is a rather important issue involved here: Since issuan-
ces are a major part of the Library's activities, the question is should
the mechanization planning be centered around these activities, or
should the planning strive for the larger goal, meaning document stor-
age ar.i retrieval and all that this implies. The first course lacks
flexibility for activities other than issuances, while the second course
implies universal adaptation and therefore requires a much more exten-
sive systems designing job, plus a great deal of foresight. I think
the choice is dictated automatically by the nature and scope of the
Library's activities, so we shall insist that the eventual system be
versatile and so constituted as to be capable of giving us new products
and services without having to make major changes to accommodate new

Here are the principal areas where computers and computer-like
devices will be employed.

Issuances. The citations for the Bibliography of Agriculture
will be punched on either paper tape or punch card, or perhaps typed
on a sheet of paper for reading by an optical scanner, such as the kind
that the Department uses in New Orleans in its cotton program. Anyway,
the citations will be recorded on magnetic tape via a computer. When
enough citations have been accumulated for an issue, they will be sorted
onto a new tape in the order in which they will appear in the Bibliogra-
phy. This magnetic tape will be converted back into paper tape which
will be used to drive a photo-composing machine to produce the printer's
copy. The engineering advances being made in the composition field will
make this conversion from one tape to another unnecessary. It will soon
be possible to use magnetic tape directly to drive a composing machine
at the speed of some 440 characters per second. The fastest that today's
paper tape can go is about 40 characters per second.

From the information supplied in the original citation the com-
puter will extract the author and subject indexes, coordinate the head-
ings with their subheadings, sort them in proper order, and print them
out. We may also want the computer to extract the names of journals
that had been indexed for that monthly issue and print them out as part
of that issue, to serve as an additional aid in searching.

The same capability will be used for the printing of special
bibliographies; of special lists, such as a complete inventory of pub-
lications that are regularly indexed in the Bibliography of Agriculture;
and of accessions list or card catalog in book form.

Document Retrieval. The mechanization of these activities is
perhaps the most responsible and crucial aspect of the entire systems
design. It is here that a most careful and virtually prophetic design-
ing job must be done. We must be sure that the activities we take care
of today will be compatible with those likely to be called for in the
future; that changes could be made without destroying what went on be-
fore; and that satellites or subsystems could be grafted onto the basic
system, calmly, unobtrusively.

Of course, what is stored and how it is stored determines what
will be retrieved. Generally speaking, we will store the data in such
a way, and so much of it, that it should be possible to demand of the
system that it meet the expected and unexpected conditions that may be
set. For example, in retrieving references on a subject the conditions
may call for a systematic retrieval for the issuance of a regular annual
bibliography on, say, growth substances. Or it may be a single, one-
time request for references on plant indicators. The specifications
may be couched in terms of languages, calling for references in English
only, or in English and French but not in Russian, or in Russian only
but only those references having summaries in English, and similar com-
binations. If we store abstracts along with the titles, then specifi-
cations may be for only those titles having abstracts. The conditions
may also be based on time periods so that only the references falling
within certain dates will be retrieved. Or it may be that the refer-
ences wanted are only those with a map, or illustrations, or having
five pages or less.

A minute ago I mentioned the grafting of satellites or subsys-
tems onto the basic system. This is a good place to mention just two
such developments because they illustrate the fact that information re-
trieval is a problem for others besides libraries.

I am sure most of you are familiar with the federally financed
research projects that are carried on at the experiment stations. I
think you have a set of cards representing each of these projects at
Gainesville. As I understand it, there are some 6,000 projects in this
category currently going on in 50 states. To these should be added an
increasing number of non-federally financed projects which are being
added to the file, as well as some 9,000 research projects administered
by the Agricultural Research Service and the Agricultural Marketing
Service. These 16,000 or so research projects are pretty well docu-

mented in terms of object or commodity classification, titles and ab-
stracts of projects, participating departments, investigators, and fund-
ing and regulatory data. The files can be approached from these view-
points with a minimum manipulation of cards. But they are not easy to
approach from the specific information retrieval angle. It would be
difficult, for example, to retrieve information on how much money is be-
ing spent on water-related research, where, and by whom. It is just
such practical information that is likely to be sought. Anyway, extract-
ing information from these projects presents about the same problem as
retrieving information from any other document. The difference is one
of emphasis rather than of substance, with an administrative detail here
and there thrown in for good measure. Both should be capable of solu-
tion by the same basic system. If it should be solved within the frame-
work of our system, then the Library would become a custodian of these
records, would maintain them, and would supply all permissive and legi-
timate requests for information from them.

The other development is a simpler one and concerns that admir-
able, literature-conscious profession forestry. Just like everybody
else, only more so, foresters have been quite concerned about keeping
up with their literature. In this case it is the literature on forest
pathology. One of our forest scientists hit upon a scheme that would
refer forest pathologists to a few widely used abstracting journals for
information they presumably need and should have. The references would
be all in English because it is felt that English is rapidly becoming
the universal language. Briefly, the scheme would work this way. As-
sume a forest pathologist wanting to know what has been written since
1950 on the subject of the fungus Fomes, or maybe its species annosus,
on pines generally, or perhaps on a species of pine. He directs his
question to the center handling the project and the center makes a
search of the records that had been stored in machine-readable form.
The results of the search are printed out by the machine, and the re-
questor is sent a list of titles of papers on the subject showing
where he can find the abstracts of these papers. He then would consult
these abstracts and decide which he wanted to pursue to the original
article and which he could safely disregard.

This scheme is still in its formative state. What will come
out of it depends on factors beyond our control. However, the Library
has offered its facilities conditionally as a possible project center.

These two developments are mentioned as an example of the care
with which the overall systems designing job must be approached, so
that both the expected and unexpected requirements could be readily

Housekeeping. And finally there are scores of routine, mechan-
ical chores that make a documentation program tick. The foremost is
the maintenance and control of records for some 25,000 outstanding or-
ders for publications that are issued periodically all over the world.
These records are constantly updated, revised, and checked for various
reasons. We are planning to mechanize these records so that repetitive
routines will be handled automatically. The machines will not only keep

these records up to date but will also print out routing cards in antic-
ipation of the arrival of publications. They will tell us regularly the
journals that should have been received but have not been; which are miss-
ing and should be claimed; and provide us with many other statistical
and signaling services, instantly and regularly. Computer-like devices
will be used to print out periodically the entire 25,000-card file in
book form. This could be done in a matter of hours as against months
that a manual job would take. As a matter of fact, the last such man-
ual job was done in 1958. This volume is badly out of date and in des-
perate need of revision, but it has been impossible to do anything about

The versatility of computer-based systems will allow us to pro-
duce equally versatile products for external information and internal
administrative purposes using the same basic file. For example, we
may want to have a listing of journals being received from Russia in
the field of forestry. This is no idle curiosity either, because, among
other uses of such inventories, one is to enable libraries to judge the
adequacy of their collections in various subject areas. The manipula-
tion of card files to produce such and similar answers and products by
hand would be prohibitive; with machines it is a matter of minutes.

Computer-like devices will be used to print catalog cards in
the form and quantity required. This is simply an operation parallel-
ing the production of card catalog in book form by the same equipment.

We will also rely on computer technology to give us a better
control of our circulation and lending activities and records.

ALL THESE AND OTHER things will come about not because they are fash-
ionable or because they bring prestige to the organization, but because
they are imperative if we are to give the kind of library service de-
manded by the times. We shall look to the technical and material wealth
of the Department to help us bring these programs to an orderly and ear-
ly fruition.

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs