Tips for searching the World Wide Web
Dozens of search facilities are available to search the web, but no single search facility is perfect. Some facilities work better with detailed queries, while others work better with broad topics. Three search engines, AltaVista, Yahoo and Lycos, exemplify a broad range of capabilities. These facilities and many others are accessible from the Internet button on the Emory home page.

Major Search Engines
AltaVista: AltaVista, with a database of more than 30 million web pages covering more than 275,000 sites, offers the most comprehensive search facility on the web. AltaVista also allows you the greatest number of options for refining your search. The drawback of such a large index is that AltaVista often returns a high percentage of irrelevant results unless you take advantage of its advanced querying capabilities. AltaVista supports the logical operators "and," "or" and "not." In addition, AltaVista supports a NEAR operator, which will return a result if two terms are within 10 words of each other. For example, an advanced query of: (William NEAR Chace) AND "James Joyce" returns three bibliographies that include works by President Chace about James Joyce.

Yahoo: Unlike AltaVista, which searches the full text of web documents, Yahoo is a hierarchically arranged, hand-edited subject index. Yahoo is most appropriate when you want to get a general idea of what is available for a given subject. For example, you should use Yahoo when you want to find the major sites about William Shakespeare, not when you want to find out about the use of mythology in A Midsummer Night's Dream.

Lycos: Lycos integrates some of the characteristics of both AltaVista and Yahoo. Lycos allows the same options for refining a query, such as using "and" and "or," but lacks the more sophisticated features of AltaVista. Like Yahoo, Lycos includes a subject index, but Lycos' subject index remains less comprehensive than Yahoo's. Lycos returns good results with minimal effort. Usually, you can type in one or two key words and expect a reasonable result without resorting to complicated querying features or lengthy traversals of hierarchical layers of links.

General Tips
Whichever search engine you use, here are several tips that will help to make your search more successful:

*Use logical operators such as "and" and "or" if possible in your query.

*If you are looking for two words that appear next to each other, such as a person's name, enclose the name in quotes. For example, a search for Thornton Dial at AltaVista returns 20,000 matches, only a handful of which are relevant; searching for "Thornton Dial," on the other hand, returns 47 matches, the first of which is the current exhibit at the Carlos Museum.

*The contents of the web change rapidly. Pages move to new locations, and searches will often lead to broken links. When you come across a broken link in a search, you can often go to a higher level in the URL returned by the search and get to the information you need. Suppose, for example, that you are searching for guidelines to web style, and a search engine returns the following broken link: http://www.emory.edu/WWW/style.html. You can remove the last part of the URL, in this case "style.html," and get to the main page for The Emory Web Project, which you can then follow to the correct URL: http://www.emory.edu/WWW/WEB_INFO/style.html

*Enter as many specific terms as possible in your search. Usually, the more terms you use, the more accurate the result.

*Try several queries at each site. Sometimes logical operators, quotation marks and additional search terms improve the results; sometimes they do not.

*Finally, always try several search facilities. The search engines employ different methods to find information and return different results. Flexibility and persistence go a long way in helping you get to the information you need.

URL's for the sites in this article include:
AltaVista http://www.altavista.digital.com/
Yahoo http://www.yahoo.com/
Lycos http://www.lycos.com/

Marie Matthews is the webmaster of Computing Resource Services in the Information Technology Division.

Return to the September 3, 1996 contents page