(Archived January 16, 1997)
| Genie Tyburski is the Research Librarian for Ballard Spahr Andrews & Ingersoll in Philadelphia, Pennsylvania and the editor of The Virtual Chase Web Site: A Research Site for Legal Professionals.
S UGGESTED READINGS ABOUT SEARCH SERVICES
Finck, Leslie, “Catch a Fleeting Moment: A Comparison of Search Engines of the Internet’s World Wide Web,” (February 8, 1996)
Gray, Terry A., “How To Search the Web: A Guide To Search Tools,” (Revised: 10/10/96)
Notess, Greg R., “Internet ‘Onesearch’ with the Mega Search Engines,” On the Nets column, Online (November 1996)
Singh, Amarendra and David Lidsky, “All-Out Search,” PC Magazine Online (December 1996)
Winship, Ian R., “World Wide Web Searching Tools: An Evaluation,” VINE (99) 1995, 49
Zorn, Peggy, Mary Emanoil, Lucy Marshall, and Mary Panek, “Advanced Searching: Tricks of the Trade,” Online (May 1996) 14
| C onducting Web research with online rationale — that is, consistently entering keywords, or so-called “natural language” queries, at search database sites to locate information on the Web — risks missing myriads of potentially useful resources. The Web is not a database; nor is it “online.” It’s a system more analogous to the size, content, and organization of several massive physical libraries than to an online system comprising countless databases.
Search database sites probe data collected by the site. Sites known as mega search sites, or mega search engines, quickly explore data gathered by several individual search database services. They do not develop their own databases. Neither search database services, nor mega search engines, store or probe all data available on the Web.
Further, many search database sites require the submission of a URL in order to collect data from an unknown location. Some do not follow all links within a site. Others, called spiders or robots, that in theory regularly collect and update data from known sites, cannot access newly discovered locations whose servers expressly prohibit it with a robots.txt file.
What strategy should one employ to initiate research on the Web? First, set aside online thinking and habits. When used with an understanding of their individual features, strengths and limitations, search database sites and mega engines often yield fruitful results. It is important to note, however, that these resources do not constitute the sole means for obtaining information on the Web.
Second, refresh traditional research skills. Imagine a large unfamiliar physical library. What would expert researchers do to find pertinent resources within the library? Most would consult a librarian or the catalog. The catalog or the librarian would in turn refer them to finding aids such as indexes, pathfinders or guides, and to treatises, online databases, CD-ROM products, or other potentially relevant resources within the library’s collection.
By analogy, the Web is a vast library, but with notable differences that effect research strategy. For example, no one manages the Web. No one evaluates the quality of resources or assesses the needs of the patrons. No one oversees weeding the collection, archiving information, or making acquisitions. While classic foes like censorship prove difficult in this environment, so does quality control.
In addition, physical libraries experience lost or damaged resources, but sources of information on the Web frequently alter content or location without record of the change. Many times information that previously existed disappears without a trace never to return to either the same location or another one.
Recognizing the need for increased sensitivity to issues of quality and authority, researchers should consider their degree of familiarity with potential resources and with the subject. In other words, in the physical world, expert researchers lacking familiarity with a library’s collection or with the subject initiate research by consulting the catalog or another expert. Those knowing the collection but uninstructed in specific resources relating to the topic seek guides, indexes, pathfinders or other finding aids. Researchers familiar with both the collection and the subject browse relevant areas of the library in search of substantive resources.
On the Web, the first scenario — unfamiliarity with both the collection and the subject –presents the greater challenge because finding a catalog may be an adventure itself. No signs nor maps point to such tools as they do in a physical library. But means for finding them exist. They comprise reading articles like this one, using known catalogs to discover others, posting questions to appropriate discussion groups, and using search services.
In closing, conducting comprehensive research on the Web requires using a variety of tools some of which closely resemble publication types researchers have consulted for years. When undertaking a new research project on the Web, consider how expert researchers might approach the same task in a physical library. Then attack it in a similar fashion allowing for differences between the two environments that effect strategy and for innovations in publication types.