Library database advisors emerging innovative augmented digital library services
Gerry McKiernan
Why should we make our users hunt down the best resource for a given information need and learn how to use its particular options for searching? Why not provide them with a simple way to get started? (Tennant, 2001, p. 29).
Readers' advisory services
A readers' advisory service has been defined as "a patron-oriented library service for adult fiction readers" (Saricks and Brown, 1997, p. 1). While there were precursors (Boone, 1996; Green, 1876), readers' advisory services initially flourished in the twentieth century in the USA in the 1920s through the 1940s (Saricks and Brown, 1997, p. 3). In his study of continuing education and the US public library, Robert Ellis Lee describes three phases of reader guidance, or readers' advisory service, between 1922 and 1940 (Lee, 1966;Saricks and Brown, 1997, pp. 3-6).
In the first phase (1922-1926), reader guidance was a special and specialized service, covering non-fiction as well as fiction, and provided separately from other library services. In this form of readers' advisory service:
Advisors met with patrons in private interviews, often in offices set aside for that purpose, to determine an appropriate reading plan. The advisors then prepared "individualized reading courses for persons who read systematically to meet the practical needs of daily living" (Saricks and Brown, 1997, pp. 3-4).
During the next phase (1927-1935), the scope and effectiveness of reader services were increased. Among the factors that led to the growth of such services was systematic research relating to problems of adult reading. Such studies provided a basis for the techniques librarians used in "linking book selection with 'the reading interests, habits, and abilities of adults', thus making readers advisory services more effective" (Saricks and Brown, 1997,pp. 3-4).
The third phase (1936-1940) saw the broadening of services in which subject specialists worked with readers' advisors in preparing reading lists. This development indicated the expansion of the readers' advisory service and highlighted the extensive reading background required to prepare individualized,annotated book lists. As demand for these specialized services increased, list preparation became more difficult (Saricks and Brown, 1997, pp. 4-5).
For librarians involved in early readers' advisory services, the role of the public library in adult continuing education was a central interest. As noted by Saricks and Brown, librarians saw a readers' advisory service as a means to assist adults in meeting their need for further learning.
… [W]hat was stressed in these early efforts was (1) the unique opportunity which libraries offered for continuous informal adult education; and(2) the "friendly and close relationship which reader's advisers formed with patrons" (Saricks and Brown, 1997, p. 5).
The aim of many involved in readers' advisory services was to "move readers toward classic works [and] to outline a plan of reading that would be educational, not recreational" fulfilling "almost a tutorial role with patrons" (Saricks and Brown, 1997, p. 6). While commendable, such highly personalized services became too burdensome. Unable to meet the demand for individually prepared and extensive reading lists, the lists became more standardized (Saricks and Brown, 1997, p. 6).
While readers' advisory services waned after the Second World War, in recent years there has been a renaissance. In 1994, the Reference Users Services Association (RUSA) of the American Library Association (ALA), for example,established a readers' advisory committee. In its first three years, the committee "presented a program on readers' advisory training …,published a bibliography of readers' advisory resources, and planned a program on readers' advisors and the Internet" for the ALA annual meeting in summer 1997 (Saricks and Brown, 1997, p. 6).
Enter the Web
With the widespread adoption and use of the Internet for information access and distribution, a number of libraries have extended traditional readers'advisory services to the World Wide Web. In that it was among the first major public libraries to offer readers' advisory services (Saricks and Brown, 1997,p. 4), it is most appropriate to find that the New York Public Library, through its branch libraries, provides one of most noteworthy compendia of readers'advisory resources. Its Electronic Reader's Advisor (http://www.nypl.org/branch/services/oas/ra.html)offers categorized and annotated "lists and links" for general fiction,genre fiction, award winners, book lists and recommendations, as well as "resources useful in reader's advisory work" for librarians. Likewise, the Inland Library System (http://www.inlandlib.org/Reference/readers.htm),a consortium of Southern California public libraries, has established a similar Web-based reader's advisory resource.
Overbooked (http://www.over booked.org/) is a megasite that specializes in literary and genre fiction that includes "author Web pages, annotated lists of nonfiction,fiction, and mystery books" with starred reviews, "themed booklists,'featured titles list and hot lists of hard cover US fiction releases', as well as 'reader's advisory resources"'.
The Morton Grove Public Library (Illinois) Webrary offers an enhanced reader's service titled MatchBook (http://www.webrary.org/rs/matchbooksearch.html),a Web-based service that provides library users with lists of new purchases"customized to each subscriber's specific interest". Using a drop-down menu, patrons are provided a listing by select subjects (e.g. "architecture","sports", "travel"), genres ("fantasy", "horror",mysteries), and/or formats (e.g. "large type (nonfiction)", "audio books", "CD-ROMs") (see Figure 1 ). A listing for a maximum of three subject groups can be generated using the MatchBook "search" form.
Figure 1. Screen print of MatchBook database search form showing drop-down menu of select subject genres or formats
Computer-based assistance
The Thompson-Nicola Regional District Library System (British Columbia,Canada), through its Reader's Robot (http://www.tnrdlib.bc.ca/rr.html),offers not only categorized lists by genre and sub-genre, but also a search system that allows a user to search a database of more than 4,500 book records by "appeal". Using this search option, users may restrict a search to one or more book genres (e.g. "mystery", "fantasy", "romance")and limit it to specific book characteristics (e.g. book length, complexity,style) or content (e.g. focus, characters, outcome) (see Figure 2).
Figure 2.Select parameter questions in Readers's Robot
For books in a variety of genres, AllReaders.com (http://www.all readers.com/) permits readers to specify a wider range of parameters for the plot, main character,main adversary, setting, "structural options", and theme. For example,using the Gordonator Precision Search Engine, a user may designate that the main character has one or more special powers, is intelligent, physically strong, and in his 40s-50s (see Figure 3).
Figure 3. Select characterization options available for searching the AllReaders.com database of "Mystery/thriller adventure and war"
In the commercial arena, EBSCO Information Services offers NoveList (http://www.epnet.com/database2.html#nov),an online, readers' advisory service that assists users in identifying fiction books based on books read or topics of interest. In NoveList, readers can identify candidate books by title or author search, or by "describing a book they would like to read". NoveList provides extensive subject and keyword access to over 100,000 fiction titles with full-text reviews and descriptions for over 56,000 books.
One of the most sophisticated readers' advisors is Book Forager (http://www.branching-out.net/forager/),a Web-based recommendation database that allows the user to select parameters using a sliding scale in a variety of categories from one extreme to another ("happy"to "sad"). As the slide is slid for a parameter, the associated icon changes to reflect the relative degree of change (e.g. a smiling smiley face to a sad smiley face). There are 20 million different individual permutations possible in Book Forager. Based on the configuration of the search parameters, users are provided with a listing of book titles that match the search parameters and domains, with links to extracts for most titles.
Database advisors
In recent years, an increasing number of technologies have been developed and applied to assist users in identifying not only books in relevant genres but online information resources as well. As with most Web-based readers' advisors,many library database advisors are categorized and annotated collections of significant Internet and other electronic resources (see for example, Fairfax County Public Library Database Advisor (http://www.co.fairfax.va.us/library/dbsRemote/dbradvisor.htm). However, a select number of database advisors are utilizing simple and advanced interfaces and technologies to assist patrons to identify relevant digital information resources.
Navigator (Yale University)
At the Yale University, David Stern, Director of the Science Libraries and Information Services, has developed the Navigator system (http://www.library.yale.edu/science/wizard.html),"intelligent front-end interfaces using nested WWW HTML screens" that direct users to the most appropriate types of sources for particular information needs (Stern, 1995) (see Figure 4).
Figure 4.Overview page of the Yale University Science Libraries Navigator
This approach begins with the type of information desired … a definition, short description, annual review, the latest information, etc.… and then moves on to subject and source determination. It assumes that users have no idea of the plethora of resources available upon entering an information maze. It attempts to identify the best and other relevant sources for a particular need.
The Navigator system contains several "expert" components:
The logical series of questions that determine which (what type) of information source is required (i.e. encyclopedia v. journal article (this emulates the accumulated knowledge of a good reference librarian)).
The subject-specialist linking of related databases within the "consider these other databases" portions of the help.
The selection of databases to be presented (filtering from among the many possible options).
The redirections that help mistaken selectors find the better/more appropriate tools for their desired information (finding journal article title words as opposed to journal name words).
Customized views for different user populations (i.e. the geology v. physics pages with differing options).
NoodleQuest
NoodleQuest (http://www.noodletools.com/noodlequest/)is a Web-based Internet search wizard developed to assist users of varying skill levels to search the Net more effectively by identifying relevant quality sites and services based upon answers to a series of seven general and broad questions(see Figure 5). Based on the selection of one or more predefined answers for each question, users are provided with a list of relevant Web sources appropriate to their needs and level of expertise that can then subsequently be queried.
Figure 5. The first three-quarters of the NoodleQuest questionnaire
Database Advisor (University of California, San Diego)
Database advisors have been developed to include recommendations for a variety of digital resources from a variety of sources. One of the first broad-based advanced database advisors was that offered by the Science and Engineering Library of the University of California, San Diego (UCSD). Within the UCSD Sciences Database Advisor (http://scilib.ucsd.edu/Proj/dba_public/),users were able to cross-search more than 25 commercial and free bibliographic databases from a variety of vendors by keyword or phrase simultaneously. Among the subject categories and the associated databases were biology (e.g. BIOSIS Previews), chemistry (e.g. METADEX), computer science (e.g. INSPEC), engineering(e.g. DOE Reports), and medicine (e.g. MEDLINE Plus). Users could search all available databases or limit a search to one or more selected subject categories(e.g. biology, chemistry, computer science, etc.) In addition, available backfiles for all or selected files could be included in the search with the current files.
Cross-search results would be ranked in descending order by database according to the number of matching records. Subsequently, users could then perform a direct search within the individual databases identified as potentially relevant for a given search request. Unlike the functionalities found within the individual databases, however, the UCSD Database Advisor recommended that users not perform author searches, phrase searchers, use wildcard or truncation symbols (e.g. "*", "?") or perform Boolean searches. A companion database (http://scilib.ucsd.edu/Proj/dba_public/dba2.html) was also developed for cross-searching more than 30 social sciences and humanities databases in a number of broad subject categories (e.g. anthropology and archaeology; art, architecture, and photography, business and economics;history; literature and linguistics; philosophy and religion; and political science and law (UCSD).
SearchLight
SearchLight (http://searchlight.cdlib.org/cgi-bin/searchlight)is a service of the California Digital Library developed from the UCSD system. Available since January 2000, SearchLight offers one-stop searching of abstracting and indexing databases, library catalogs, and Web sites, as well as other types of resources in two broad subject domains, the sciences and engineering, and the social sciences and humanities. In SearchLight:
The user's search words are sent to a wide variety of databases (well over 100), with the results organized by resource type (books, journal indexes,electronic journals, e-texts and documents, reference resources, and web directories). The number of hits is noted beside each resource, and clicking on that number will automatically take users to the results in that particular database when possible. Alternatively, they can click on a link to go to the resource and search it directly. Anyone can try it out, but those not part of the UC community won't see results for licensed databases (Tennant, 2001, p. 29).
As with the UCSD Database Advisor, within SearchLight users can search all available databases in a broad domain, or limit a search to one or more specific subject categories (e.g. biological and medical sciences, engineering and technology, physical sciences).
Article database advisor (University of Illinois)
Californian universities are not alone in the development of advanced tools to assist library users in the identification of appropriate electronic resources. Since 1998, librarians at the University of Illinois have been involved in the investigation and development of a "Web-based expert system tool that could assist users in identifying the most appropriate online or CD-ROM database to meet a particular information need" (Ma and Cole, 2000a,pp. 2-3). Known as the UIUC Smart Database Selector Project, the effort sought to:
identify and better understand computer-assisted techniques for guiding users to the databases most suitable for their particular information needs;
pioneer innovative methods to improve user services given the proliferation of electronic databases and digital information within individual institutions;and
improve the ease of access to information by taking advantage of the Web and related protocols and technologies (Ma and Cole, 2000a, p. 3).
To achieve these objectives, the project "created a selection database containing rich and extensive characterizations of the available resources… [R]esource character-izations were developed using database documentation, standard reference sources, information supplied by librarians most knowledgeable about the subject area of each resource, and whenever available, the controlled vocabularies of the resources themselves" (Ma and Cole, 2000a, pp. 4-5). Overall, the process was intended "to mimic the process a librarian uses when selecting a database to search for a particular information need" (Ma and Cole, 2000a, p. 5). The key component of the Smart Database Selector was the inclusion of the controlled vocabulary terms used by most of the databases characterized (Ma and Cole, 2000b).
The Database Selector Project has been implemented as the Article Database Advisor (http://gateway.library.uiuc.edu/rex/selector.htm),an experimental database advisory system.
SUMSearch™ (University of Texas)
Another of the most innovative locally developed library database advisors is SUMSearch™ (http://sum search.uthscsa.edu/searchform45.htm), a service of the University of Texas Health Science Center at San Antonio (Badgett et al., 2001). SUMSearch™ is a "unique method of searching for medical evidence"using the Internet by combining meta-searching and contingency searching. Search results are organized by breadth of discussion (see Figure 6):
Figure 6. Sample search results in SUMSearch™
First, there are links to resources that provide broad discussion:relevant textbooks, followed by traditional review articles, and practice guidelines. Next there are links to resources that provide narrow discussions:systematic reviews, and original research. Thus, the clinician that is searching a topic with which they are not familiar, will find links to easy to read, broad discussions at the top of the list. A clinician that has a specific question within a topic with which they are otherwise familiar, will find links to systematic reviews and original research in the second half of the results.
Users may "focus" a SOMSearch to one or more subject aspects (e.g."treatment", "diagnosis, prognosis", etc.), or age or language. In addition, search results can be limited to one of three specific source "textbooks"(i.e. Merck Manual, Canadian Task Force on Preventive Health, AIDS Knowledge Base).
FlashPoint (Los Alamos National Laboratory)
Another special library that offers amalgamated searching of a variety of electronic resources is that affiliated with the Los Alamos National Laboratory(LANL) in New Mexico. Among its range of resources are BIOSIS Previews,arXiv.org e-print server, INSPEC, the local online catalog, electronic journal collections, and the science and social sciences citation indexes. With FlashPoint (http://flashpoint.lanl.gov)– a multi-database search tool – registered users can cross-search any or all of these locally mounted resources simultaneously. FlashPoint is a component of the Library Without Walls Project of the LANL Research library(Mahoney and Di Giacomo, 2001).
Multi-Search (University of Arizona)
Using the SiteSearch software from OCLC, the University of Arizona offers Multi-Search(http://www.library.arizona.edu/indexes/links/multisearch.shtml), a cross-database searching service that allows campus and registered users "to search for books and articles in different catalogs and indexes at the same time". Users can restrict a search to a specific topic or specify the individual indexes and catalogs to be searched. Within Multi-Search, "SiteSearch uses the Z39.50 protocol to search databases that are compliant with that standard – in this case, three state catalogs and the OCLC FirstSearch databases" (Tennant, 2001, p. 29).
Westchester Library System Gateway
The Westchester Library System gateway (http://www.museglobal.com/try/)offers access to a variety of electronic resources of interest to the member libraries and patrons of this New York State county library cooperative. Its"Quick Search" option allows users to simultaneously search electronic resources from selected broad categories by free-text terms or phrases, while its "Advanced Search" option allows users to limit a search to keyword,title, author, subject, or "description" fields and select specific candidate sources.
Target resources searchable within the gateway include full-text newspaper and magazine sources, abstracts and indexes databases, and reference databases. Among the sources accessible and searchable from within the gateway are Business and Company Resource Center, ERIC, Facts on File, FISOnline, Funk & Wagnalls New World Encyclopedia, Middle Search and Primary Search. Upon initiating a Quick Search, the selected sources will be systematically searched and available results displayed for each.
Unlike the service previously noted, the Westchester Library System uses Muse, commercial software produced by MuseGlobal (http://www.museglobal.com/). Muse is not a search engine but a "multiprotocol universal search environment". Muse technology is unique in that it combines the broadcast search ability of a meta-search engine with a structured searching ability for simultaneous access to library catalogs, proprietary full-text databases, document repositories, Web engines and resources, digital infobases, and more.
KCLS Library Search (King County Library (Washington State)
The King County Library System (http://www.kcls.org)of Washington State is another library that utilizes commercial software to enable its patrons to cross-search its local OPAC, Web site and more than 60 databases. Its KCLS Library Search utilizes WebFeat (http://www.webfeat.org), a cross-platform,cross-search system. The Brooklyn Public Library, Houston Public Library, and San Francisco Public Library are among the other major public libraries, and other organizations, that have adopted WebFeat to provide one-stop searching of electronic resources for patrons.
Technologies
WebFeat is but one of an increasing number of commercial software systems that have been developed to facilitate the identification and use of the diversity of electronic resources. Endeavor Information Systems (http://www.endinfosys.com), the producer of the Voyager integrated online library system, recently introduced ENCompass (http://encompass.endinfosys.com),a "framework that allows libraries to offer a single search interface to a variety of information resources" such as:
a local library catalog (OPAC);
commercial databases;
electronic journals;
electronic books;
digital archival collections;
Z39.50 resources;
audio or video multimedia collection;
locally-developed image databases; and
remote OPACs.
ENCompass provides end-users with "unified access to all of …[a] library's resources, eliminating user confusion as to which interface to search". Its "architecture allows libraries to integrate all of their collections, regardless of location or format", allowing users to explore multiple collections to identify the most relevant information to their information needs. Among the libraries that have implemented ENCompass are Kansas State University, Auburn University, the University of Pennsylvania, and the national Library of New Zealand (Endeavor Information Systems, 2001).
Readers' Database Advisor
More than 70 years ago, S.R. Ranganathan, the noted librarian and library educator, published the first edition of his seminal work, The Five Laws of Library Science (Ranganathan, 1931). These five laws concisely represent the ideal service and organizational philosophy of most libraries today:
- 1.
books are for use;
- 2.
every reader his book;
- 3.
every book its reader;
- 4.
save the time of the reader;
- 5.
a library is a growing organism (Garfield, 1984, pp. 6-7).
Such laws are not only applicable to librarianship in general but characterize the establishment, evolution, and enhancement of readers' advisory services and library database advisors as well. From their initial inception and implementation, readers' advisory services have sought to link users with relevant books and other information materials, and books with users. With the proliferation of digital resources in recent years, libraries and librarians have begun to facilitate access and use of the diversity of electronic resources available from a variety of vendors using advanced database advisors, truly"saving the time of the reader" and fulfilling the intent of this principle. In view of current trends, it is clear that digital resources will become further integrated and accessible. Such integration, however, will not be limited to selected resources; through collaborative filtering and other emerging digital technologies, a readers' database advisory service that integrates community interaction with digital collections will no doubt arise. Such an enhancement will clearly symbolize the continued development of the library as both a place and service, further realizing the fifth of Ranganathan's law.
Acknowledgements
The author is grateful to the following individuals for granting permission to use screen images from their respective Web sites.
ReferencesBadgett, R.G.,Paukert, J. and Levy, L. (2001), "The evolution of SUMsearch™ for teaching clinical informatics to third-year medical students", Academic Medicine, Vol. 76 No. 5, p. 541.Boone, A.L. (1996), "Readers'advisory service in the public library: past, present, and future", a Master's paper submitted to the faculty of the School of Information and Library Science of the University of North Carolina at Chapel Hill in partial fulfillment of the requirements for the degree of Master of Science in Library Science, July 1996.Endeavor Information Systems (2001), "Kansas State University: an ENCompass case study", available at: http://www.endinfosys.com/customers/ksu.pdf (accessed 8 February 2002).Garfield, E. (1984), "Current comments: a tribute to S.R. Ranganathan, the father of Indian Library Science. Part 1. Life and works", Current Contents, No. 6, (February 6), pp. 5-12, available at: http://www.garfield.library.upenn.edu/essays/v7p037y1984.pdf(accessed 8 February 2002).Green, S.S. (1876), "Personal relations between librarians and readers", American Library Journal, Vol. 1 No. 2-3, pp. 74-81), available at: http://dlis.gseis.ucla.edu/people/jrichardson/personal.htm(accessed 20 January 2002).Lee, R.E. (1966), Continuing Education for Adults through the American Public Library, 1833-1964, American Library Association, Chicago, IL.Ma, W. and Cole, T. (2000a), "Creating an electronic databases selection expert system", paper presented at the Illinois Association of College and Research Libraries Spring Conference, Hot Topics 2000, held in Matteson, Illinois, April 13-14, 2000, available at: http://dli.grainger.uiuc.edu/Publications/TWCole/IACRL_DBSelector/DatabaseSelectionExpertSystem.pdf(accessed 6 February 2002).Ma, W. and Cole, T. (2000b), "Genesis of an electronic database expert system", Reference Services Review, Vol. 28 No. 3, pp. 207-22.Mahoney, D. and Di Giacomo, M. (2001), "Flashpoint@ LANL.gov: a simple smart search interface", issues in Science and Technology Librarianship, No. 31, Summer, available at: http://www.library.ucsb.edu/istl/01-summer/index.html(accessed 7 February 2002).Ranganathan, S.R. (1931), The Five Laws of Library Science, Madras Library Association, Madras.Saricks, J.G. and Brown, N. (1997), Readers' Advisory Service in the Public Library,American Library Association, Chicago, and London.Stern, D. (1995), "Expert systems: HTML, the WWW, and the librarian", Computers in Libraries,Vol. 15 No. 4, pp. 56-8.Tennant, R. (2001), "Cross-database searching", Library Journal, Vol. 126 No. 17, October 15, pp. 29-30, available at: http://libraryjournal.reviewsnews.com/index.asp?layout=articlePrint&articleID=CA170458(accessed 3 February 2002).
Gerry McKiernan(gerrymck@ iastate.edu) is a Science and Technology Librarian and Bibliographer, Iowa State University Library, Ames,Iowa.
