SUBSCRIBE NOW!

Vol. 38 No. 5 — June 2021

IN THE BEGINNING

Well before the internet, the web, and commercial search engines such as Google, organizations started to develop ways to digitize data and write instructions for searching that data. For instance, in 1966, Chemical Abstracts Service (CAS) started using automated methods to improve the production of its printed products, which then led to spinning off computer-readable magnetic tapes that could be searched using programs of instructions. Initially, magnetic tape for data storage was wound on 10.5-inch reels. This standard for large computer systems persisted through the late 1980s. In 1968, CAS made commercially available a database on tape, called CA Condensates, comprising bibliographic information and keywords.

In a fortuitous set of circumstances, in 1966, the Chemistry Library at Indiana University (IU)–Bloomington was completely renovated and enlarged, and John Knego was hired as head. It also happened that the nonprofit Aerospace Research Applications Center (ARAC) was established in Bloomington in 1962 as a joint effort by NASA, IU, and private industrial firms. The bulk of ARAC’s collection consisted of thousands of technical reports from NASA, the Atomic Energy Commission, and other governmental entities from which it offered selective dissemination services to industrial customers. Prior to 1965, only the NASA database was searched by ARAC, but then it started to acquire tapes from the Library of Congress, Engineering Index, and, in 1968, CA Condensates. ARAC developed a standard file format program that used a weighted term strategy for searching these databases and employed a staff of trained chemists to work with users to develop search profiles. Creating custom profiles was time-intensive because a too-narrow number of search terms would produce few or no citations, whereas a too-broad one would produce a large number of irrelevant hits. Searches needed to be run a few times to reach the proper degree of specificity. By 1968, ARAC started developing standard interest profiles with topics that would satisfy small groups of researchers on specific topics.

Coincidentally, in 1963, Harry Shull, a member of IU’s chemistry faculty, start- ed a service called the Quantum Chemistry Program Exchange (QCPE) in a Quonset hut next to the IU chemistry building. Its purpose was to facilitate the exchange of software. In 1967, it hired Dick Counts, one of the developers of the ARAC search program, to be its administrator.

BIRTH OF THE CHEMICAL INFORMATION CENTER

The added responsibilities and services of the Chemistry Library led John Knego to receive permission to expand his staff. Since he knew me as a faculty wife who had majored in chemistry, and he was familiar with papers I had written in library school at IU, he offered me a position in 1972. His expectation was that I would work full time, but since I had small children, I was unwilling to accept. Therefore, he hired another full-time librarian, Peggy Green, as his assistant, and he put me in charge of the newly created Chemical Information Center (CIC) on a half-time basis. The CIC took over responsibility for the selective dissemination of information (SDI) service, based on CA Condensates, which ARAC had been providing.

The operation of the batch SDI service consisted of creating, maintaining, and making changes to more than 135 interest profiles, which were kept on IBM punched cards (stiff paper stock containing 12 rows of 80 columns); receiving the weekly magnetic tape of CA Condensates; and taking the tape, punched profiles, and search program to the university computing center. From 1970 to the mid-1980s, most computer access was via punched cards, which are now obsolete. Programs and data were punched by hand on a keypunch machine and were read into a card reader. The program instructions were written in the FORTRAN computer language. After the program was run on a mainframe computer (originally a CDC 3600), the results were printed out oncontinuous pin feed computer paper. The output was then separated into the individual profiles, trimmed, and distributed.

This system worked well, and although the results were hardly elegant, and changes could only be made weekly, not interactively, users were mainly satisfied. I enjoyed the work, but carrying boxes of cards, tapes, and bulky printouts weekly to and from the computing center a few blocks away was rather laborious, so I had a graduate student, Paul Shullenberger, help with these duties. My slightly larger problem was when there was a glitch in the process at some point. Since cards ran to several hundred, and it was important that they be read sequentially, there was always the chance for something to go wrong. Initially, I had Dick Counts help me with programming problems, but, eventually, I was pretty much on my own. Consequently, I found it necessary to learn some FORTRAN. On one occasion, at the beginning of a volume year at CAS, some minor format changes to the tape caused the program to fail to run. I had to follow the logic of the printed roughly 2,000 lines of code instruction by instruction for a couple of days before I could figure out the problem (which was really quite simple in the end) and recompile the program.

ONLINE ARRIVES

While this batch service was ongoing, John Knego became aware of a new system called ORBIT, created by the System Development Corp. (SDC), a spinoff of the RAND Corp. in California. In 1970, SDC started a service called AIM-TWX that provided access to the National Library of Medicine’s (NLM) Abridged Index Medicus (AIM) through telex to hospitals. Ann Van Camp, the medical librarian at the IU School of Medicine, did searches using this service and could be called the first “online” searcher in Indiana. In 1973, SDC commercially offered a version of CA Condensates that could be accessed from California interactively using a portable computer terminal. The CIC became the first ORBIT customer in the state of Indiana, and, therefore, I may have been the second “online” searcher in the state. The terminal we used was the Silent 700, which was introduced in 1971 by Texas Instruments. It printed with a 5 x 7 dot-matrix heating element onto a roll of heat-sensitive paper. It was equipped with an integrated acoustic coupler and modem that could receive data at 30 characters per second. We later subscribed to a competing service called DIALOG, a subsidiary of Lockheed Martin, also in California. Besides CA Condensates, databases mounted on the systems included MEDLINE, Education Resources Information Center (ERIC), BIOSIS, EMBASE, and PsycINFO.

There were numerous growing pains with the early use of these systems due to the limited equipment, transmission problems, and computer reliability. I recall that on one occasion, a scheduled DIALOG presentation had to be canceled because the computer in California was down. I learned early on that it was not sensible to enter lengthy Boolean search phrases because of transmission problems. Instead, it was best to enter single terms and combine later. The output on thermal paper was far from pleasing. Later, the Silent 700 was replaced by a video terminal with an electronic cathode ray tube for display instead of paper. To this was added a printer and a modem to provide for printed output and communications through telephone lines. This equipment was used throughout the remainder of the 1970s until the age of personal computers began with the introduction of the first IBM personal computer in 1981.

ONLINE SEARCHING EXPANDS

In 1976, John Knego left for another library position at Florida Atlantic University and was replaced by Gary Wiggins as head of the Chemistry Library and director of the CIC. About 1979, I became a full-time tenure-track employee, and in my first year, I performed 334 online searches. In 1978, the IU Main Library established the Computer Assisted Reference Service (CARS) using an OCLC terminal with two searchers for the Bloomington campus and one for regional campus queries. The CIC online search service was integrated into CARS officially in 1980. Over time, additional branch libraries were added to the service, and more librarians were trained in search techniques. Initially, databases were searched exclusively using index terms, but it soon became possible to search titles, abstracts, and more as well. Gary Wiggins expanded the Chemical Information Specialist Program that John Knego had developed with the IU School of Library Science. One of the requirements of the program was to help students master the techniques of SDI profiling and learn the structure of the CA Search database. I taught the portions of the course devoted to creating profiles.

Following ARAC’s lead, the CIC also started developing standard interest profiles that could be sold commercially to help recoup the costs of its services. A standard interest profile is a current awareness publication in a narrow subject area appearing at regular intervals. By then, a number of other organizations were also producing standard interest profiles, including the BioSciences Information Service, CAS, the Institute for Scientific Information, and the QCPE. In London in 1973, P.S. Hunter published An Index to Computer-Produced Standard Interest Profiles in Chemistry, Applied Chemistry, Chemical Engineering and Metallurgy. In 1979, I became the editor of An Index to Standard Interest Profiles in Science and Technology, published by the CIC. This was intended as an update and expansion of Hunter’s work. A second edition was produced in 1981.

In fall 1981, I left the CIC to temporarily run IU’s Swain Hall Library (catering to physics, mathematics, astronomy, and computer science) under the direction of Gary Wiggins, while there was a search for a permanent head. I continued to do online searching, and, in 1982, when online searching was still relatively new, a group of four Hoosier librarians—Meri Meredith (Cummins Engines), Pat Riesenman (CARS), Becki Whitaker (Indiana Cooperative Library Services Authority; INCOLSA), and I—founded the Indiana Online Library Users Group (IOLUG). INCOLSA is now the Midwest Collaborative for Library Services. IOLUG is still in existence nearly 40 years later, although it has expanded to include all current trends in libraries and technology.

In 1983, I left IU to work as a technical information librarian for AT&T in Indianapolis, eventually becoming supervisor of the Technical Library. While there, I learned about many new technologies just then developing, including cellphones, compact discs, telefax, HDTV, and smart homes. The AT&T Indianapolis location started undergoing major downsizing in 1986, with reductions in space, budget, and personnel, so job security was not guaranteed, especially if one was not willing to transfer to another location. Therefore, I applied and was hired for an opening as head of the Medical Sciences and Optometry Libraries back at IU.

END-USER MEDLINE

Having now become primarily a medical librarian, my searching emphasis turned to the MEDLINE database from NLM. I attended Medical Library Association training and eventually met the requirements to become a member of the Academy of Health Information Professionals. MEDLINE is highly structured and initially required extensive training. However, during the 1980s, databases—including MEDLINE—offered full-text searching versions of their software, which made information more easily retrievable. Widespread use of personal computers led to user-friendly front-end systems for end users of the information, bypassing the need for specialist searchers. There were a few different versions for MEDLINE, the best-known being GRATEFUL MED from NLM. Compact Cambridge also introduced MEDLINE on CD-ROM for end-user searching. CD-ROM databases became popular in the mid-1980s, especially in reference departments of libraries. They required special CD-ROM workstations but eliminated the online connection charges associated with traditional online searching.

My career took another turn when I was hired by the NutraSweet Co. in Deerfield, Ill., in November 1989 as scientific information specialist, where I remained until my retirement in 2000. While there, I was responsible for online searching, scanning standard interest profiles, obtaining articles (usually by fax), maintaining a small library, and keeping and processing copies of virtually all printed materials in the field of sweeteners.

THE INTERNET AND BEYOND

The 1990s saw the birth of the internet on a commercial scale and the introduction of search engines such as Yahoo and Google. The freewheeling search capabilities (even by voice recognition) of these systems make it possible to retrieve information of any quality quite easily, as there are no longer gatekeepers to ensure the reliability of information. However, for serious scientific work, there is still a need for information professionals and end-user training.

After retirement, I continued to do searching as a volunteer in Northwestern Memorial Hospital’s Health Learning Center for 14 years, until the pandemic halted volunteer services in March 2020. The center subscribed to publishers’ services that made it possible for us to deliver not only bibliographic citations, but also the full texts of articles to the desks of doctors and nurses—as well as do customized searches for all users of the center and provide training for end users.

It has been an extremely eventful 50 years, with amazing advances in technology, but also questions concerning reliability of information, cost of access, copyright issues, effective search strategies, and the training of users. I can’t even imagine what the next 50 years will bring!

BIBLIOGRAPHY

Association of Indiana University Chemists NEWSLETTER. Chemistry Library and Chemical Information Center Provide a Variety of Services for Teaching and Research. V. XXIII. p. 5-6. November 1977.

Bjørner, Suzanne. Online Before the Internet, Part I: Early Pioneers Tell Their Stories. Searcher V. 11 No 6, June 2003. Part 2. Searcher 11(7) July/August 2003.

Bonham, Miriam D. An Index to Standard Interest Profiles in Science and Technology. Chemical Information Center, Department of Chemistry, Indiana University. 1979, 1981.

Bonham, Miriam. The Indiana University Chemistry Department’s New Chemical Information Center. The Accelerator. June 1973.

Bonham, Miriam D., Woodhouse, Laurie L. An Evaluation of Four End-User Systems for Searching MEDLINE. Bulletin of the Medical Library Association. 76(1) January 1988.

Buntrock, Robert E. Searching Chemical Abstracts vs. CA Condensates. Journal of Chemical Information and Computer Sciences. 15(3) 1975.

Capodagli, James A., Mardikian, Jackie, Uva, Peter A. MEDLINE on Compact Disc: End-User Searching on Compact Cambridge. Bulletin of the Medical Library Association. 76(2) April 1988.

Computer Search Center, IIT Research Institute. Chicago, Illinois. Searching Chemical Abstracts Condensates, On-Line and Batch. Journal of Chemical Information and Computer Sciences. 16(3) 1976.

DiSalvo, Joseph, Hall, Robert W. Standard Interest Profiles: Development of Technical Subjects. ARAC, Final Report. Bloomington, February 1968.

Hunter, P.S. An Index to Computer-Produced Standard Interest Profiles in Chemistry, Applied Chemistry, Chemical Engineering and Metallurgy. London, Polytechnic of North London, School of Librarianship, 1973.

Prewitt, Barbara G. Searching the Chemical Abstracts Condensates Data Base via Two On-Line Systems. Journal of Chemical Information and Computer Sciences. 15(3) 1975.

Riesenman, C. Patricia. Reference Use of Computers. Indiana Libraries. 1(1) Spring 1981. p. 26-30.

Roberts, Anita B., Hartnell, Ieva O., Counts, Richard W., Davila, Roberta. A. Development of a Computerized Current Awareness Service Using Chemical Abstracts Condensates. Journal of Chemical Documentation. 12(4):221-223. November 1972.

Wiggins, Gary. The Indiana University Chemical Information Specialist Program. Science & Technology Libraries. 1(3) Spring 1981.