How Not to Waste Knowledge

The library that is most familiar to people is a hushed, well-stacked place, commanded--at least in fable--by a hushed, not-very-well-stacked spinster with her hair in a bun and ice in her eyes, and frequented by those who want to browse or drowse, study or doodle, read or simply exchange furtive notes with members of the opposite sex. There are still hundreds of such cozy havens all over the U.S., but they are turning into anachronisms. Their problem is that a technological age demands far more from a library than a quiet place to read and a random assortment of books.

Casual users of libraries are hardly aware of it, but library professionals and their more conscientious clients know about it all too well. They call it the "information explosion," and it has precipitated an odd paradox: most of the nation's public libraries have neither the money to buy nor the space to house the books and periodicals that a growing and insatiable public wants to read, while the technical disciplines--chiefly the sciences--have turned loose such a Niagara of information that even the wealthiest of corporate, collegiate or community libraries simply do not know what to do with it, let alone how to make it available to researchers.

"National Disgrace." Both the information explosion and the population explosion have forced libraries of all kinds to expand and to build anew at a spectacular rate, often with striking esthetic effect (see color pages). U.S. colleges alone more than doubled their annual library-building outlay, from $21 million to $58 million, between 1957 and 1962, and spent a total of $211 million. In the succeeding six-year period, from 1963 through 1968, they will have tripled that amount to $650 million. The nation's largest library, the Library of Congress, has just renovated its main reading room at a cost of $309,215.

Still, the nation's public libraries are in bad shape. The American Library Association, which sets a minimum standard of 10,000 library volumes for communities with fewer than 2,500 people, figures that 69% of the 7,260 public libraries in the U.S. are substandard. Of the 8,000 elementary schools in the country, fully 60% have no central libraries, a state of affairs that U.S. Education Commissioner Francis Keppel calls "a national disgrace."

And, despite all the expansion, Allan Cartter, vice president of the American Council on Education, reckons that only 17% of the nation's college libraries meet the 100,000-volume standard that is considered minimum for good undergraduate instruction. Only 25 graduate schools, moreover, can boast the 1.5 million volumes considered minimal by the council. In all, says Cartter, only two dozen academic libraries are "really adequate." Among the best: Harvard's libraries (7,245,000 volumes), followed by those of Yale (4,703,000), Illinois (3,748,000), the University of California at Berkeley (2,956,000), Cornell (2,577,000), Stanford (2,416,000), and U.C.L.A. (2,007,000).

Those schools, however, are exceptions. As Cartter says, there is a close correlation between the size of a library and the quality of a university. Thus most U.S. colleges are severely handicapped by inadequate libraries, since, among other factors, it is the quality of the library that attracts good faculty members and graduate students.

Acres of Cards. Unless even the best schools find ways of keeping current with new information, they too will run into incalculable trouble. Last year alone, 20,542 new books and 7,909 new editions of older books were published in the U.S.--nearly twice the number that came out in 1960. In addition, 22,262 periodicals and 80,000 technical reports rained off the presses. In many cases, it would have been merciful to save good paper. At the same time, there is no question that a lot of worthy material will go wasted and unread.

Yet who knows what is being missed? About 90% of all scientists who ever lived are now at work--and, it seems, most are publishing their findings. In 1750, there were about ten scientific journals in the world; today there are about 7,000 related to the biomedical sciences alone. Once scientists wrote about physics, chemistry and biology; today they deal with the likes of biochemistry, bioengineering, exobiology and biophysics. In 1950, chemists produced 558 articles every two weeks for their publications; in 1965, in the field of chemistry alone, those learned explorers are turning out--and publishing --6,700 articles every fortnight. Small wonder that the U.S. Printing Office is drawing up plans for a new building with 40 acres of working space--six acres bigger than the Pentagon; or that Yale, if it were to continue using its obsolescent card catalogue, would need eight acres of floor space by the year 2040 just for the cards alone. The books would be virtually unhousable.

Skatinq in the Stacks. For specialized libraries in particular, the only answer to suffocation by paper is automation, which runs from the crudest idea to the most dazzling concept of the sophisticated mind. In the Detroit Public Library, book procurement is speeded up by sending boys whizzing down the 250-ft.-long stacks on roller skates. On the other hand, scientists dream that one day a scholar will be able to quiz a regional computer by telephone from his office; whereupon the answer, perhaps from a paper by a foreign colleague, will bounce off an orbiting communications satellite first into a simultaneous translator and then on to the scholar's TV screen.

Between these two extremes, much has already been achieved. Dozens of libraries are using data-processing machines to record book purchases, to keep track of the books that are lent, and even to grind out overdue notices. But that kind of automation mainly helps the librarian. More significant automation is aimed at helping the reader and researcher discover precisely what information is available. Uncounted millions of dollars are wasted annually by scientists repeating research that someone else has already painstakingly carried out and published. An odd medical fact tucked away in a periodical might save a life if the right doctor only knew that it was there. So far, however, only a fourth of the nation's available scientific literature is now catalogued. Many physicists are aware, for example, that one of the most complete collections of Chinese physics periodicals rests in an Air Force research laboratory at Massachusetts' Hanscom Field, but due to the lack of an adequate cataloguing system, none really know what is in it.

Apart from the problem of translation, the technology of bibliographic control in scientific fields is already a reality. A leader in the field is the National Library of Health in Bethesda, Md., which tries to acquire every publication relating to medicine. Librarians feed selected references from articles in 2,400 periodicals into two Honeywell computers. Then, by the use of key words, the computers each year arrange 150,000 citations alphabetically. This list is printed by a computer-driven phototypesetter, and the result is a book, the Index Medicus, which goes to 7,000 libraries around the world. A researcher in London, for example, can leaf through the Index, find the citation he wants, and request it of his local librarian, or, if necessary, seek a copy from Bethesda. Chemists benefit from a similar service in Columbus, Ohio, where articles on their specialty are abstracted and indexed by computer; the "express-indexes" are sold by subscription. Bookless Colleges. The next step after bibliographical control is retrieval of information itself. Computers cannot yet actually "read" documents and pull out the relevant parts. They can, however, select copies of entire documents. The most advanced example of this technique is the CIA's "Walnut" system, by which an 8-in. by 14-in. page of information is reduced to a microscopic image; any one of 990,000 images can be plucked out of the CIA's computers within five seconds, after which it can either be projected on a screen or reproduced on paper.

Academic libraries are laying plans to tie themselves together electronically so that each can benefit from the resources of the others. The medical libraries of Harvard, Yale and Columbia have been preparing punch cards for three years for such a network. Another system is Project Intercom, shared by a steadily growing number of colleges (now ten), which hope to unite their literature through computers, beginning with the medical and biological sciences. Eventually, says Intercom Executive Director Dr. James Miller of the University of Michigan, there will be a network of bookless libraries--study booths, electric typewriters and TV screens--in which "a small college will be able to have a better library than Harvard has now."

Blow to Browsers. The real key to that kind of progress is automation of the recently facelifted Library of Congress, which tries to catalogue every publication in the U.S., now has 40 million 3-by-5 cards in its files. Converting that catalogue to tapes for computer use would cost about $40 million--but a librarian anywhere in the U.S. would then have a constantly updated index to all U.S. publications. The Library of Congress could use a Data-Phone system to flash information by ordinary telephone line to any library that requests it. The system, already employed for high-speed communications, uses perforated tape that transmits signals to remote printing machines at the rate of 6,000 words per minute--all for the price of a long-distance telephone call. The only fear is that large-scale mechanization of libraries may dampen the joy of the ordinary reader, who simply likes to mull through the library shelves. And it could be a big blow to serendipity in the stacks: gone will be the excitement of discovering the right answer in an unexpected place. On the other hand, automation will put a higher premium on the professional ability of librarians, who will have to organize knowledge far more precisely than the Dewey Decimal System demands. "All the money in the world isn't going to get a computer to judge what is worth storing and what is not," says the University of Pennsylvania Assistant Director Jesse Mills. Amid the proliferation of paper, that kind of judgment will continue to be far more valuable than any technical breakthroughs.

This file is automatically generated by a robot program, so reader's discretion is required.