Nicholson Baker - Double Fold - Libraries and the Assault on Paper

Clapp’s last and most ambitious mechanization scheme — an attempt to create a working electronic library at MIT — was called Project Intrex. In the early sixties, Intrex (administered by Carl Overhage and other veterans of air-defense engineering at MIT’s Lincoln Laboratory) had been sponsored by the Independence Foundation, a conduit for CIA money 61 61in that era; in 1967, the Council on Library Resources assumed lead financial responsibility. Clapp took a personal interest in Intrex, serving on its steering committee even after his retirement from the presidency of the Council; the project envisioned, among other things, “better and more economical systems for weeding,” 62 62as well as “digital storage of encoded full-text 63 63in massive random-acccess storage.” (“Massive” and “mass” were thermonuclear words that seemed to get the hearts of information scientists beating faster.) But Intrex’s historian, Colin Burke, sums up the project thus:

Project INTREX fell very short

64

64

of the expectations of all its sponsors. After some eight years INTREX ended with little more than a few pieces of soon outdated hardware, some homeless software, and twenty thousand indexed articles in a limited field called “material science.”

Intrex’s only visible achievement, Burke adds, was a set of paper finding aids called “Pathfinders,” which helped students get around the reference collection.

Why couldn’t Clapp have shown a little patience, and funded more quiet inquiries into techniques of cataloging Persian works (as some traditionalist members of his board 65 65would have much preferred him to do), reconciling himself to the fact that whatever glorious man-machine couplings were in the offing, they weren’t going to happen in his lifetime? Why couldn’t he have left library administrators alone, rather than forever distracting them from their primary task as paper-keepers by dangling the lure of convulsive change before them, long before the change was practical, and long before it had revealed its many risks? Clapp especially goaded his alma mater, the Library of Congress, to invest in seductive prematurities, early systems that broke down, cost a fortune, spread confusion, didn’t focus, made life more difficult, and failed in general to do what they were built or bought to do. Forty years and many generations of scrapped prototypes later, libraries are still trying to get Clapp’s remote-access full-text wish-list to fly.

But brute shrinkage was the idea closest to Clapp’s heart, and there (despite the disappointments of machines like the Verac) the basic technology was already mature. Microtext, he wrote, has “rescued many millions of pages 66 66of newspapers from oblivion at comparatively low cost and with a concomitant saving of space”—the next step, then, was to perform a similar sort of “rescue” on journals and then books. The question was how to pay for it. In College and Research Libraries, Alan Pritsker and William Sadler had, in 1957, dealt an inadvertent blow to the burgeoning micro-movement in a cost study that found microfilming to be of financial appeal as an alternative form of storage only if libraries (and their patrons) were willing to tolerate (1) somewhat fuzzy print, (2) minimal quality control, and (3) “the destruction of the text.” 67 67(The researchers assumed textual destruction because the microfilming system that they were using in their estimates of work-flow, the RemRand Model 12, required that pages be “fed automatically into the machine.” That in itself was not a problem, the authors wrote: “Since the purpose of microfilming is to reduce the space requirements, the cutting of the bindings is considered inconsequential. Any possible gain from the resale value of these books would be more than offset by the increased efficiency in filming.”) And yet, despite their diligent efforts, Pritsker and Sadler found microfilm conversion to be a costlier means of storage for research libraries than bookshelves. (Alan Pritsker, incidentally, went on to become a pioneer of digital simulation, creating the computer languages JASP for the Air Force and GERT for NASA.)

Clapp responded to these unwelcome results by having the Council commission a similar study, completed in 1961. The Crerar Library 68 68in Chicago, a privately endowed reference library whose social sciences collection had been sold off by Clapp’s former Library of Congress colleague Herman Henkle, was supporting itself by marketing its services to places like the Atomic Energy Commission, whose Nuclear Science Abstracts Crerar’s librarians produced. The library was moving 69 69from an old building to a squat glass box on the campus of the Illinois Institute of Technology; there were many pre-1920s journals in the collection which took up space. Clapp and Henkle felt, or hoped, that microfilming these volumes might be cheaper than continuing to store them, but they couldn’t be sure, especially with Pritsker’s discouraging results; so Clapp asked a small consulting firm to perform an operations-research analysis on the problem, “Costs and Material Handling 70 70Problems in Miniaturizing 100,000 Volumes of Bound Periodicals.” Clapp, too, allowed his engineers to assume “shearing of spines” in their estimates, since spine-shearing allows for “considerable labor saving 71 71in the photographic operation by avoiding the necessity for raising and lowering the pressure plate each time a page is turned.”

This time around, the cost comparisons came out a little better. It now appeared, according to Clapp’s summary of the research, that with enough buyers of prints of the microfilm, a large microfilming project could successfully reduce a library’s storage costs without any of the sacrifices that Pritsker and Sadler enumerated — no sacrifices, that is, “except that of destruction 72 72of the text.”

The newspapers went first, but as the filmable remainder of their own bound backfiles dwindled, library planners began to look around for other ways to occupy their now fully staffed and equipped information-renewal programs. “It’s like having a sausage factory, in a way,” one former Library of Congress department head told me. “You’ve got to feed the beast.” (The library owned twenty-four microfilm cameras 1in 1973; they were shooting seven thousand feet of negative film per day.) Books with brittle paper were one good possibility — shabby, unattractively aging, toned by time. In the mid-sixties, the library, again in the vanguard, began segregating thousands of books that were (as the 1968 annual report of the Council on Library Resources phrased it) “otherwise beyond redemption.” 2Coincidentally, the library needed more space: “Space was a key word 3in the thinking and activities of this division [the Office of Collections Maintenance and Preservation] during fiscal 1966,” reported the Library of Congress Information Bulletin. In 1967, Verner Clapp’s last year as president, the Council gave the Library of Congress (via the Association of Research Libraries) a grant for a Pilot Preservation Project, to explore “arrangements for assuring the preservation 4of these [brittle] books for the continuing uses of the research community.”

One interesting idea, which had been propounded by Gordon Williams in a 1962 study (also prepared with the help of a grant from the Council), was to save a physical copy of every significant book in a central, low-temperature storage warehouse, where it would be available for microfilming on demand. A benefit of the plan, according to Williams, was that libraries would then know what they could “safely discard” 5if they wanted to, since there was one backup in deep freeze. This well-intentioned conceit proved in the end administratively unwieldy, but the research done in connection with it, which compared some of the Library of Congress’s books with the same titles held in other libraries, showed that the condition of a given book “varied greatly” 6from library to library: an important observation, since it implied that book longevity depends on local variables (humidity and temperature, rough treatment, styles of rebinding) as much as it does on the innate chemical properties — the “inherent vice”—of the paper.