Frederick G. Kilgour

Pioneer of inter-library cataloguing

Frederick Gridley Kilgour, librarian: born Springfield, Massachusetts 6 January 1914; President and Executive Director, Online Computer Library Center (OCLC) 1967-80; Distinguished Research Professor, University of North Carolina 1990-2004 (Emeritus); married 1940 Eleanor Beach (three daughters); died Chapel Hill, North Carolina 31 July 2006.

When the history of libraries in the 20th century comes to be written, there will be a special place in it for Frederick G. Kilgour, who in 1967 cut the Gordian knot of cataloguing. Librarians had up to then insisted on the quality of the information in the description of books; Kilgour showed that quantity could be just as important, and proved the point by building a co-operative network that now stretches worldwide.

Computers, Eniac in particular, had been one of the factors that won the Second World War for the Allies. The war over, a new breed of computerologists looked round for new worlds to conquer. They built systems to deal with the vast detail of government departments. They then went on to the very few commercial giants with a payroll, customer-base or inventory so large as to be worth the then huge cost and complex machinery involved in computerisation. That done, they looked round for something else, something that existed in vast quantities and needed sorting, alphabetically or numerically.

Their eye lit on libraries - books, millions of them, that all had to be sorted by author, title, subject, shelf-mark, a vast new world to conquer. So they descended on libraries, crying, "Boys, you can throw away all those slips and cards and typewriters. We can do it all for you." "Oh," said the librarians, "so you think cataloguing books is easy. Let us just show you a thing or two," and they proceeded to show how much experience and knowledge it took only to record all those details that looked so simple, let alone sort them in any order.

Nothing daunted, the computerologists sat down and invented systems that met what the librarians said were the basic standards of cataloguing. The librarians, interested in spite of themselves, began to grapple with what the computerologists called "logic", which struck the librarians as the very reverse of what the philosophers on their shelves said it was. Just like Stickly-Prickly and Slow-and-Solid in the Just So Stories, they came to grow more and more like each other, and the armadillo of the computer-cataloguer came into existence.

The Library of Congress, the temple of book-cataloguing in the United States, had been the epicentre of this process. For years, it had created, and issued to other libraries in letter-press printed form, cards that fitted into cabinets in libraries all over the country. Were these to be supplanted by electronic data? It was not going to be easy. Few libraries could afford the equipment, none had experienced staff; above all, cataloguing itself, up to the standards imposed by the Library of Congress, was so difficult. Enter Fred Kilgour.

Born in Massachusetts, Kilgour had gone to Harvard and found a job in its library before the war. Joining the US Navy, like many librarians he gravitated into intelligence work, specialising in the collection and dissemination of information acquired from enemy sources. On leaving government service in 1948, he went to Yale, where he became associate librarian for research and development. Then, in 1967, he was lured away by the Ohio College Association to run a system that would pool the catalogues of its 54 libraries in the state.

OCLC, the "Ohio College Library Center", was born. The task before it was, in theory, massive. Each of the libraries had records of their holdings, many of them copies of the same book, all in slightly different forms. They could be, and were, transferred into electronic data, but there they showed up as different books. It would have taken an eternity to standardise them. Kilgour took the common-sense decision: it didn't matter - all readers in libraries needed to know was that a book something like what they wanted existed, and where it was.

The success of this experiment, launched in 1971, was immediate. From its base in Dublin, Ohio, OCLC took in records from any library that wanted to participate. As computer cataloguing improved, so did OCLC's systems. Funny things happened from time to time, as when a whole lot of manuscript records crept in by mistake, but by and large it got better and better, though never rising to the Library of Congress approved standards of RLIN (the Research Libraries Information Network).

But, for the ordinary needs of ordinary libraries, OCLC did very well. The deal was simple. OCLC charged new members so much a record for supplying digital records of any book they had on file; if the new entrant could provide details of any book not on OCLC, it paid them. Its initials now stand for Online Computer Library Center, and it has some 10,000 participating libraries all over the world (among them our own London Library), and more than a billion records. Very soon, all these will be available online to anyone with an internet connection.

Kilgour retired in 1980, and was latterly Distinguished Research Professor Emeritus at the School of Information and Library Science at the University of North Carolina, where he went on teaching until he was 90. He knew and wrote a lot, about the history of science and the evolution of the book, but nothing was quite as important as that flash of inspiration, almost 40 years ago, that brought OCLC into being.

Nicolas Barker