|
|
|||
|
|
|
|
|
|
||||
        
|
 |
Chapter 9: Search
Full Web Searching Overview
The requirements for Webwide searching are daunting. Users expect to be able to quickly type in a simple search phrase at a global search engine like AltaVista or Google and end up with a realistic result. Consider the chances of walking into a public library and finding a particular passage in a book in a few seconds and you'll understand the near futility of the task. When searching, users are often overwhelmed with too much information, are shown irrelevant information, or do not get anything at all. Despite the resulting frustration, users keep pounding away at search engines, hoping to get a good result in a matter of minutes.Many of the problems with search engine usage have to do with users not searching correctly. Searching really should be used only when looking for known items or for very specific topics. Consider searching for a general term like "hamburgers." Search engines may not necessarily pull up sites about hamburgers or even large hamburger restaurant chains. In fact, testing this query in some search engines resulted in numerous links to pages about Hamburg, Germany, as well as recipe sites for personal home pages and pages that appeared to have absolutely nothing to do with hamburgers. The problem is that the search term isn't specific enough. If you search for something like "White Castle Slyders"—a regionally famous hamburger in the United States—you may find a more useful list of results. When looking for general information on a subject, users usually turn to a directory rather than a search engine. The main difference between a search engine and a directory is that a directory usually involves some human editing and usually contains a very limited number of links. Yahoo! is probably the most famous directory around, but it now provides search engine features as well. In fact, most of the search engines have begun to offer directory links as well as searching. Many popular search engines now focus more on delivering users to sites that focus on a particular topic rather than trying to get users to a very specific site. Directories like www.about.com or www.dmoz.org are organized by individuals who are responsible for a particular type of content. The benefit of a directory is that having a site organized by people can result in limiting content to just the "good sites." While automatic gathering and categorization of content can be a powerful tool, until Artificial Intelligence is vastly improved the value of human editing and categorization should not be underestimated. Definition: A Web directory is a human-edited and organized collection of site links and associated information such as descriptions and reviews.
In comparison to a directory, a pure search engine is more like the phone book that you can only search. This is similar to calling your information service and asking for a phone number, except you ask for something related to a particular topic. Consider using a phone information service such as 411 in the United States and asking for the phone number of a "Chinese restaurant" rather than asking about a particular Chinese restaurant. If you ask for a particular restaurant, chances are you're going to get a good result. However, when asking for general information you'll be very lucky if the operator actually spends some time to give you a particular restaurant they know about, or even returns one that looks reputable based upon its ad in the print directory. In many cases, directory assistance might just give you the first one or even a random one from a list. Search engines tend to act the same way. They are good at returning specific answers, but results vary otherwise.Search engines always attempt to be comprehensive and may list numerous sites without regard to content quality or freshness. Search engines are primarily automated in the collection and organization of links, though today some human editing as well as directory-oriented results are being used. This is due partly to the massive amount of search engine trickery going on, as well as a desire to improve the result sets for users. The reason for the trickery is a desire by Web site owners to use search sites to drive as much traffic to their sites as possible. Next: Search Engine Promotion
|
 Overview | Chapters | Examples | Resources | Buy the Book! |