First-Hand:The Birth of MELVYL
Contributed by Stephen R. Salmon
MELVYL was the first online library catalog. It was developed, beginning in 1977, at the University of California's Universitywide Library Automation Program (ULAP) in the Office of the President (then called Systemwide Administration).
The development of MELVYL was part of a much larger effort. UC was under pressure from the State budget office to increase cooperation among the nine campus libraries, and the University brought me in to coordinate that effort, and also take charge of the library automation program. I wanted to create a "union" catalog that would enable students, faculty and staff to see what was in the University's nine campus library collections as a whole. There was an existing group called the Universitywide Library Automation Program (ULAP) that had produced a book-form catalog, much criticized for its many mistakes. I hired some very smart technical people, and we began to think it might be possible to develop an online union catalog. We included it as part of the master plan I developed ("The University of California Libraries: A Plan for Development," University of Californis, l977), and I was able to convince the legislature to fund it the following year (1978), even though that was the year of Proposition 13 and a lot of budget-cutting was going on.
The prototype system became available for use by staff members in 1980, and was made available to library users the following year.
The name "MELVYL," by the way, was an inside joke. In the initial version of the system, we used ORVYL and WYLBUR, the time-sharing software developed at the Stanford Linear Accelerator Center (SLAC). So I used the same naming convention (changing an "i" to a "y") and name our system after Melvil Dewey, the most well-known innovator in library science.
The development of MELVYL naturally got a lot of attention from other libraries, who wanted an online catalog, too, so in 1981 I founded a commercial company (Carlyle Systems) to develop such systems. Rather than the mainframes approach used in the UC system, we used multiple processors of our own design and a proprietary database system. We had the first commercially-available online library catalog. Our first customer was the largest research library in the country, outside of the Library of Congress: the New York Public Library. Exciting times!