A New Frontier: The Impact of a 1959 Board Meeting

Guest blog by Ken Koyle, MA, Deputy Chief of the History of Medicine Division (HMD) at the NIH National Library of Medicine. This post celebrates the important work performed by our archival professionals and the archival collections held by the library, from which the source material was drawn, as NLM celebrates International Archives Week #IAW2022.

In November 1959, when construction of NLM’s current building at NIH was still underway and digital computing was in its infancy, the NLM Board of Regents convened on the third floor of the Old Red Brick building for a demonstration of the indexing process. When Board Chairman Michael E. DeBakey, MD, asked if computer technology could be used in indexing, NLM Director Col. Frank B. Rogers, MD, was ready with an answer. Dr. Rogers, clearly interested in the emerging technology of automated data processing (ADP), described an article by Robert S. Ledley, DDS, in that month’s issue of Science and noted that Dr. Ledley was already contracted with NLM to report on using computers in indexing.

Dr. Rogers was instrumental in NLM’s first explorations of automated processes and had a clear vision of the potential of electronic computing, including how it could improve efficiency at NLM, but his optimism was tempered by prescient realism. Dr. Rogers recognized—and conveyed to the Board—that the potential benefits of ADP would require a commensurate investment of staff time and labor. “We should not forget that ‘automatically’ means ‘because we told it to do so beforehand,’ and this in itself may turn out to be quite a trick.” Dr. Rogers made it clear that the computer age would bring a change in work, but not necessarily a reduction in work. “Remarkable as the capacity of the computer may be for sustaining a long sequence of operations, it is nevertheless ultimately only the end-phase of that still longer sequence which must include as a first phase the human labor of input.”

Acknowledging the upfront labor investment in ADP was only part of Dr. Rogers’ insight. He also explained that the human work was not only substantial and necessary, but also incredibly complex: “The instructions [for a computer] are a thousand times more detailed, for the simplest task, than those required to be given to the . . . clerk.” Unleashing computers’ potential would require staff to think in new ways, conceive new methods of organizing data, and embark on a new journey of continuous learning and professional development.

Along with the challenges of training staff to work with ADP equipment came the interminable problem of cost. Much as today’s public institutions are grappling with the costs of cloud computing, digitization, and increasing storage requirements, Dr. Rogers had to balance the potential benefits with the considerable costs of computer equipment. The type of computer necessary to realize Dr. Rogers’ vision would cost about $1.5 million in 1960—98% of NLM’s total budget of $1,566,000.

Undeterred, Dr. Rogers found an answer to the funding problem by collaborating with another agency that would benefit from the increased processing speed of scientific literature that the envisioned system could provide: the National Heart Institute. They provided the initial funding, NLM did the legwork, and in 1963, the new MEDLARS computer went into service. Dr. Rogers had realized his vision of bringing automated indexing to NLM. As Surgeon General Luther Terry said at the Board meeting in April 1961, “If any institution ever stood on the borderland of a new frontier it is the National Library of Medicine.”

Dr. Rogers was very clear about the issues of cost, labor, and expectations in his 1960 presentation to the Board, including his overarching concern about balancing NLM’s core mission with these potential new directions:

[The] purpose of the Library is not to operate a particular machine system, however great an acrobatic achievement that might be in itself. It is not to publish and distribute a particular index in a particular way, however ingenious and successful that operation may be deemed to be. It is not even just to be a good library, however great and distinguished that library may be. It is rather, by virtue of being a library, to use every available bibliothecal means to promote awareness of and access to the subject content of recorded medical knowledge, to the end that the science of medicine will advance and prosper.

More than 60 years later, NLM still holds fast to that purpose. As stated in our statutory mission and reiterated in our current strategic plan, we are here “to assist the advancement of medical and related sciences and to aid in the dissemination and exchange of scientific and other information important to the progress of medicine and to the public health.” Our continued pioneering work in data science is just one way we accomplish that mission.