ACRL News Issue (B) of College & Research Libraries January 1989 / 7 Library research in the future: A progn ostication By Philip H. Young Director, Krannert Memorial Library University o f Indianapolis One librarian’s speculative bibliographic scenario. L ib r a r ia n s , like other professionals, should try to predict the future course of their profession. For academic librarians especially, one im portant as­ pect of prognostication concerns how a researcher of the future will access information and use it in the creation of a new document. Let’s do some “blue-sky” dream ing about w hat th a t future might be like. The research environment will be comfortable, quiet, and conducive to efficient brain function­ ing. The furniture will be ergonomically pleasing and the surroundings free of disruptions. In this comfortable setting will be located the tools of re­ search: information sources, means to record and excerpt im portant passages, and a tabula rasa upon which to create new syntheses, ideas, and prom ul­ gations. Appropriate processing tools will be read­ ily available for th e m anipulation of d ata and words and the eventual perfection of a new docu­ ment. Most im portant for the researcher is access to the materials th at he/she needs to consult and digest while studying a topic of interest. To achieve this goal there will be a computer-driven research sta­ tion with the capability of linking him or her either to the information sources directly or to remote sources which will provide them quickly. The re­ searcher will be able to communicate w ith this ter­ minal (for lack of a better word) as easily and natu­ rally as possible, i.e., by voice. The term inal will itself be an “intelligent” machine w ith the ability to learn the researcher’s interests as the work pro­ gresses and suggest possible a ltern ate research routes along the way. The term inal’s access to a w ide array of information resources and strategies will be transparent to the researcher so as not to im­ pede his or her thought processes. The terminal will be able to shift between information searching and its other functions w ith o u t pause and, if needed, perform various functions at once so that, for example, the researcher could be reviewing an information source at the same time as composing notes inspired by it. The computer terminal will be the linking agent between the researcher and various information resources and between the researcher and a fin­ ished product, be it notes for future reference or a document for publication. The initial interchange between researcher and terminal will resemble a reference interview w ith the researcher stating the topic of interest and the terminal discovering w hat scope and type of information is needed. The ter­ minal will then make connections to appropriate resources and search for information on the re­ searcher’s subject of interest. D uring this process, the terminal may have to check w ith the researcher for clarification about the research topic in order to narrow or expand the search and to report initial findings in order to check w hether search strategies in process are on the right track. Materials in all formats will be recovered, if the researcher wants them , and citations will appear clearly and con­ 8 / C&RL News cisely (without abbreviations, code terms or num ­ bers, tight spacing, etc.) W hen the initial information search is complete and useful items or citations to them are recovered, the researcher will be able to proceed directly to work. Remotely held materials will be autom ati­ cally ordered and notices as to delivery time will be reported. Materials available online will be dis­ played immediately or printed out on a connected printer, if so preferred. While perusing online m a­ terials, the researcher will be able by voice com­ m unication to move sentences, p aragraphs, or other selections into a holding area and reorganize them or add his or her own comments and observa­ tions as the research progresses. Footnotes as to sources will be generated automatically so th at the researcher will know whence information items derived but will not get bogged down w ith the chore of remembering to ask for citations for each item. The researcher will be able to mark certain elements of his or her notes for special recall or sort­ ing later. Finally, the researcher will be able to review and organize the information recovered, along w ith his or her personal insights and developments and will be able to use these notes to build a final report doc­ ument. The notes will be able to be sorted and re­ sorted as needed w ith source footnotes always fol­ lowing and accessible. The notes will be able to be m anipulated alongside the new document being created so th at both are visible at the same time. Useful tools, such as thesauri, dictionaries, fact and date lists, etc., will be available for instant online consultation as needed. All formatting and other word-processing functions will be handled auto­ matically or by voice command if defaults are not adequate. It may be hard to visualize how the process out­ lined above might operate, so let’s imagine a re­ search session of the future. Our patron, Nancy Bookwright, enters an available library research carrel assigned to her at the front desk, closes the soundproof door to the room, and switches on the computer term inal’s power supply switch. As she sits down in the comfortable chair facing the large screens built into the desk top, the center screen dis­ plays the lib ra ry ’s logo, and a soft co m p u ter­ generated voice says, “Good afternoon, are you ready to begin an information search?” After N an­ cy’s affirmative reply, the voice requests her to read a standardized sentence aloud so th at the terminal can adjust its voice-recognition circuits for her accent and individual pronunciation. Then, the voice says, “W hat topic would you like to search for today?” “I am interested in the battle of M arathon,” she replies, w atching the almost instantaneous appear­ ance of the words on the screen. A split second later, a short encyclopedia paragraph appears be­ low the request. “Are you referring to the battle between the an­ cient Greeks and the Persians fought in 490 B.C.?” queries the term inal’s voice. After Nancy’s “yes,” the voice continues, “Do you need general infor­ m ation or specific research-level materials on this topic?” “I need to review all the research done in the last fifty years,” says Nancy, leaning back in her chair. “I am beginning the search now ,” responds the term inal as its screen displays the words “Search in Process.” D uring this interchange, the term inal has al­ ready formulated a search strategy and accessed a massive database composed of all terms used in ti­ tles and abstracts of virtually all periodical litera­ ture for the past decade. While accessing this retro­ spective periodicals database and the national book catalog, the term inal presents citations and ab­ stracts of three articles from the initial search. “Here are three recent articles th at seem to be on your topic. While I am accessing earlier records, would you like to view the full text or have a print­ out?” Nancy skims the titles and abstracts and notices th a t one of the articles is about the price battles of M arathon Oil Company w ith a competitor and th a t she has already read another one of the arti­ cles. “Please show the full text of citation tw o,” she replies. A new screen is activated and the text of the requested article is presented on the right side of the desktop display. Meanwhile, a citation appears on the left side and the term inal’s voice says, “I have only found one book dealing w ith your topic. The search term is not in the title, but the table of con­ tents refers to it.” Below the citation, the table of contents has now appeared on the screen. The voice continues. “This book is presently in the li­ brary. Would you like to have it delivered?” “Yes, please,” responds Nancy. While the term i­ nal is signaling a library page working in the appro­ priate section of the stacks to find the book and de­ liver it to the w indow slot in Nancy’s carrel, our researcher is reading the article on the right side of the screen. “Save the third paragraph,” she says, and the terminal adds a section below the scrolling text where the paragraph and its source citation ap­ pear. O ther sections are saved and personal com­ ments added as our researcher works through the article. W ithin a few minutes, the book slot beside the carrel’s door swings open and the requested book appears on the shelf there, already checked out to her by the term inal’s communication w ith the circulation computer. Nancy reaches for the book and begins to read it, again making verbal notes as she proceeds. The term inal’s pleasant voice interrupts: “There are five retrospective articles on your topic in the tim e range you specified. Only tw o of them have full text available online. Of the other three, the li­ b rary holds hard copy of two of them and I have ordered them for delivery to you. Would you like to request the th ird one on In te rlib ra ry Telefac­ simile?” After Nancy’s affirmative response, the voice continues, “For the tw o available online, January 1989 / 9 would you like to view the full text or have print­ outs?” Nancy requests printouts because she is still busy with the book source, and a high-speed, quiet printer whirs and deposits several sheets in the bin near Nancy’s chair. Well, let’s leave our patron of the future at work and return to the present to analyze this vision of future information access and scholarly research. The work place which we have envisioned is cer­ tainly nothing which could not be available now. I have situated it in a library, but other than quick delivery of materials not available online there is no reason that it could not be located anywhere, such as in the researcher’s private office or even at home. The details of comfortable chair and quiet atmosphere were included to make the point that present libraries tend to feel sterile and institu­ tional. Noise from people passing or talking can be bothersome to researchers, and the lack of privacy, uncomfortable furniture, poor lighting, etc., de­ crease their comfort. Central to our vision of the future is the terminal through which our researcher accesses information sources, conducts her research, incorporates her own opinions, analyses, and observations, and cre­ ates a final product. I have endowed the imaginary terminal w ith intelligent voice communication ca­ pability, although the above exchanges could have occurred on-screen just as well. Computer experts are working on voice recognition systems for word processing right now and have prototypes already operational. C urrent research emphasis is being placed on artificial intelligence and expert systems for computers, and the vision of a terminal with rather sophisticated voice interaction and “learn­ ing” capability does not seem to be too far down the road. Our hypothetical term inal’s other functions are, again, logical combinations of or expansions on presently available technology or systems. After the terminal has determined w hat search terms and qualifiers are needed for Nancy’s search, it proceeds to access appropriate periodicals data­ bases, perhaps more comprehensive than our BRS or DIALOG but similar to them. The terminal then accesses a national book database not unlike OCLC or RLIN but w ith the capability to review tables of contents, a development already under discussion today. Finally, the term inal accesses the holdings of its own library to check w hether the de­ sired citations match locally held items. Again, on­ line catalogs are not new, but the autom atic link­ age with search services would be innovative. The terminal’s ability to perform all these tasks simulta­ neously represents vast memory expansion over present computers, as well as elaborate artificial intelligence structures to direct activities and out­ put. However, the trend towards lower memory cost and the potential advent of new storage tech­ nologies make me think such hypotheses not too farfetched. Most im portant to our vision of future research is the rapid document delivery imagined. The con­ cept of having a terminal go beyond telling a re­ searcher th at the library owns a book to having it page someone to get it off the shelf and deliver it to a c e rta in lo catio n seems only an extension of present online catalog capabilities. Similarly, the ability to request an item recovered as a citation through an extension of the initial search using rapid library to library telefacsimile service is a com bination of p resent technical capabilities. Probably the most novel aspect of our imaginary researcher’s docum ent recovery is the ability to view (or download and print) full texts of journal articles. H owever, w ith the already noted ad ­ vances being m ade in computer memory and stor­ age technologies, perhaps the day will come when journals will either be scanned digitally for online availability or will simply have their texts distrib­ uted and m aintained electronically. W here w ould this picture of future research leave libraries? I believe the in-depth and well- packaged collections of knowledge th at we know as books will not disappear or be transformed into electronic storage anytime in the near future due to the immense am ount of data they make available, especially retrospectively. This observation may also hold for periodicals, but I think, as indicated in the above scenario, th at it is more likely that this medium may become electronic in the not too dis­ ta n t future. Thus, libraries will still be needed as repositories of information in hard copy w ith the requisite needs to organize it and make it available as efficiently as possible. Instead of people brows­ ing the stacks, we may have them browsing elec­ tronically, but they will still need the document to be delivered as quickly as possible. Libraries will need to reorganize their thinking towards deliver­ ing books to patrons using research stations in the building or in rem ote locations. Librarians will still need to select and collect books and non-book materials for their patrons’ use and they will have to develop efficient means of moving documents from library to library for swift delivery of items not held. Since the agent of access will be computer-driven terminals, programs of in­ struction on their use will be required, as well as more general instruction programs on the nature of information itself and w hat formats and presenta­ tions of it are likely to be available. In the scenario presented above, the researcher had already de­ fined a very specific topic of interest but, in fact, m any information seekers will require the assist­ ance of librarians to help them focus their needs sufficiently to utilize th e intelligent term in al’s search capabilities. Most im portantly, librarians will have to oversee and manage the complex com­ munications systems and linkages to information providers which will be the life-blood of the intelli­ gent terminals’ ability to match researcher w ith documents. In the future, as now, a researcher will w ant to locate materials of interest, get the documents in 10 / C&RL News hand, and study them as efficiently as possible in th e most co m fo rtab le su rro u n d in g s, w ith the fewest interruptions, and the least “red tap e.” The librarian of the future will be using new equipment and amazing computer linkages of data manage­ m ent systems b u t will still have the same mission as in the past: to collect, organize, and make avail­ able the information needed by researchers; to pro­ vide appropriate equipment to access, service, and update th at information; and to train patrons to use information sources efficiently and effectively. ■ ■ O ptical disk-based p erio d ica l in d exes for u n d ergrad u ates By Lori E. Buchanan User Education Librarian Austin Peay State University Anne May Berwind Head, Inform ation Services D epartm ent Austin Peay State University and Don Carlin Information Services Librarian Austin Peay State University Local library coverage and success rate as criteria fo r selecting an index. Faced with a growing number of choices and limited budgets, academic libraries must carefully consider the goals they wish to achieve w ith optical disk-based periodical indexes. The initial invest­ m ent is especially im portant for two reasons. First, the product chosen introduces the campus commu­ nity to the technology. Second, it sets the stage for future optical disk acquisitions. The goals for each library are different. In one library, optical disks are seen as an alternative to online database search­ ing. In another library, they are viewed as an excit­ ing, new way to introduce lower-division under­ graduates to the research process. Once a library sets its goals, the optical disk products likely to meet these goals are selected for evaluation. Several re­ cent articles which outline the evaluation or selec­ tion of optical disk products are found in the library l i t e r a t u r e .1 These articles provide good back­ ground information for libraries considering the purchase of an optical disk-based periodical index. However, they do not document an evaluation pro- 1Notable examples are: David C. Miller, “Evalu­ ating CD-ROMS: To Buy or W hat to Buy?” Data­ base, June 1987, pp. 36-42; Gail T. Graves, L aura G. H arper, and Beth F. King, “Planning for CD- ROM in the Reference D epartm ent,” College and Research Libraries News, July/August 1987, pp. 393-400; L inda Stew art, “Evaluation C riteria: Picking CD-ROMS for Public Use,” American L i­ braries, October 1987, pp. 738-40.