Searching the World Wide Web: Overview
Searching the World Wide Web can be both beneficial and frustrating. You may find vast amounts of information, or you may not find the kinds of information you're looking for. Searching online will provide you with a wealth of information, but not all of it will be useful or of the highest quality.
The World Wide Web is a superb resource, but it doesn't contain all the information that you can find at a library or through library online resources. Don't expect to limit your search to what is on the Internet, and don't expect search engines to find everything that is on the Web.
Studies of search engine usage show that search engines are increasing exponentially in their indexing of new Web sites and information. Indexing is the Web term for finding and including new Web pages and other media in search results. For example, in 1994, Google indexed approximately 20 million pages. As of 2004, that number is up to 8 billion! However, search engines still only index a fraction of what is available on the Internet and not all of it is up to date. Search engines may only "crawl" sites (or revisit them for purposes of indexing) every month or so; information that has been updated since that time will be invisible to the search engines. After you try several search engines, you will see that you get different results from different sites. Also, remember that some information appears and then disappears from Web sites. Finally, search engines don't always search the entire page; if a page is larger than 100 to 500 k, many search engines will only index the first 100 to 500k of the page. So there could be valuable information that is being overlooked by a search engine even in pages that are indexed.
Not all of the information located on the Internet can be found via search engines. Researchers Chris Sherman and Gary Price call this information the "invisible Web" (another name that is frequently used is the "deep Web"). Invisible Web information includes certain file formats, information contained in databases, and pages omitted from search engines.
So, using search engines is not the only way to find material on the Web, but these search engines are one tool you can use. Knowing a few search strategies and hints can make the search more profitable. This guide provides information on the different ways of locating material on the Web including using search engines, searching the invisible Web, and using Web directories.
How the Internet and Search Engines Work
The Internet is made up of a vast amount of computers networked throughout the world via data lines or wireless routers. New computers and Web sites are added every day, and no larger organizational system exists to document and catalogue them all. The Internet is a dynamic, growing, and changing system, which makes navigating it or searching it thoroughly difficult.
This is where search engines and Web directories come in. Search engines, such as Google or Yahoo, are large databases of information that store and retrieve relevant website results based on keywords. Web directories, such as the Open Directory Project, are attempts to organize the best of the existing Web sites into categories and subcategories. No search engine or Web directory will have the same sites listed in the same order, and none will have all of the possible sites on the Internet listed. Furthermore, the ranking of a Web site within a search engine (i.e. how high up on the results list it appears) has as much to do with politics as it does with quality information. The search engine rankings are determined by a number of factors including the amount of information on the site, the amount of other sites that link to it, the number of people who select that link when searching, the length of time that the site has been listed in the search engine database, and the code of the site.
Recently, search engines such as Google and Yahoo have also been providing "sponsored links"—links that appear on the first few pages of the search results and that are paid for by advertisers. This means that you may end up clicking on something that is not relevant to your search, but instead actually advertising. The image presented here gives you an example of this on Google.
What does this mean for a researcher? Understanding the nature of the Internet, how to navigate it, and how it is organized can help you assess the quality of information and Web sites, filtering out that which does not relate or is of questionalbe quality.
Kinds of Search Engines and Directories
Web directories (also known as indexes, Web indexes or catalogues) are broken down into categories and subcategories and are good for broad searches of established sites. For example, if you are looking for information on the environment but are not sure how to phrase a potential topic on holes in the ozone, you could try browsing through the Open Directory Project's categories. In its "Science" category, there is a subcategory of "Environment" that has over twenty subcategories listed. One of those subcategories is "Global Change" and this includes the "Ozone Layer" category. The "Ozone Layer" category has over twenty-five references, including a FAQ site. Those references can help you determine the key terms to use for a more focused search.
Search engines ask for keywords or phrases and then search the Web for results. Some search engines look only through page titles and headers. Others look through documents, using Google, which can search PDFs. Many search engines now include some directory categories as well (such as Yahoo).
These (e.g., Dogpile, Mamma, and Metacrawler) search other search engines and often search smaller, less well-known search engines and specialized sites. These search engines are good for doing large, sweeping searches of what information is out there.
A few negatives are associated with metasearch engines. First, most metasearch engines will only let you search basic terms, so there are no Boolean operators or advanced search options. Second, many metasearch engines pull from pay-per-click advertisers, so the results you get may primarily be paid advertising and not the most valid results on the Web.
Searching with a Search Engine
A search engine is a device that sends out inquiries to sites on the Web and catalogs any Web site it encounters, without evaluating it. Methods of inquiry differ from search engine to search engine, so the results reported by each one will also differ. Search engines maintain an incredibly large number of sites in their archives, so you must limit your search terms in order to avoid becoming overwhelmed by an unmanageable number of responses.
Search engines are good for finding sources for well-defined topics. Typing in a general term such as "education" or "Shakespeare" will bring back far too many results, but by narrowing your topic, you can get the kind (and amount) of information that you need.
- Go to Google (a search engine)
- Type in a general term ("education")
- Add modifiers to further define and narrow your topic ("rural education Indiana")
- Be as specific as you can ("rural education Indiana elementary school")
- Submit your search.
Adjust your search based upon the number of responses you receive (if you get too few responses, submit a more general search; if you get too many, add more modifiers).
Learn how the search engine works
Read the instructions and FAQs located on the search engine to learn how that particular site works. Each search engine is slightly different, and a few minutes learning how to use the site properly will save you large amounts of time and prevent useless searching.
Each search engine has different advantages. Google is one of the largest search engines, followed closely by MSN and Yahoo. This means that these three search engines will search a larger portion of the Internet than other search engines. Lycos allows you to search by region, language, and date. Ask allows you to phrase your search terms in the form of a question. It is wise to search through multiple search engines to find the most available information.
Select your terms carefully
Using inexact terms or terms that are too general will cause you problems. If your terms are too broad or general, the search engine may not process them. Search engines are programmed with various lists of words the designers determined to be so general that a search would turn up hundreds of thousands of references. Check the search engine to see if it has a list of such stopwords. One stopword, for example, is "computers." Some search engines allow you to search stopwords with a specific code (for Google, entering a "+" before the word allows you to search for it).
If your early searches turn up too many references, try searching some relevant ones to find more specific or exact terms. You can start combining these specific terms with NOT (see the section on Boolean operators below) when you see which terms come up in references that are not relevant to your topic. In other words, keep refining your search as you learn more about the terms.
You can also try to make your terms more precise by checking the online catalog of a library. For example, check THOR+, the Purdue University Library online catalog, and try their subject word search. Or try searching the term in the online databases in the library.
Most search engines now have "Advanced Search" features. These features allow you to use Boolean operators (below) as well as specify other details like date, language, or file type.
Know Boolean operators
Most search engines allow you to combine terms with words (referred to as Boolean operators) such as "and," "or," or "not." Knowing how to use these terms is very important for a successful search. Most search engines will allow you to apply the Boolean operators in an "advanced search" option.
AND is the most useful and most important term. It tells the search engine to find your first word AND your second word or term. AND can, however, cause problems, especially when you use it with phrases or two terms that are each broad in themselves or likely to appear together in other contexts.
For example, if you'd like information about the basketball team Chicago Bulls and type in "Chicago AND Bulls," you will get references to Chicago and to bulls. Since Chicago is the center of a large meat packing industry, many of the references will be about this since it is likely that "Chicago" and "bull" will appear in many of the references relating to the meat-packing industry.
Use OR when a key term may appear in two different ways.
For example, if you want information on sudden infant death syndrome, try "sudden infant death syndrome OR SIDS."
OR is not always a helpful term because you may find too many combinations with OR. For example, if you want information on the American economy and you type in "American OR economy," you will get thousands of references to documents containing the word "American" and thousands of unrelated ones with the word "economy."
NEAR is a term that can only be used on some search engines, and it can be very useful. It tells the search engine to find documents with both words but only when they appear near each other, usually within a few words.
For example, suppose you were looking for information on mobile homes, almost every site has a notice to "click here to return to the home page." Since "home" appears on so many sites, the search engine will report references to sites with the word "mobile" and "click here to return to the home page" since both terms appear on the page. Using NEAR would eliminate that problem.
NOT tells the search engine to find a reference that contains one term but not the other. This is useful when a term refers to multiple concepts.
For example, if you are working on an informative paper on eagles, you may encounter a host of Web sites that discuss the football team the Philadelphia Eagles, instead. To omit the football team from your search results, you could search for "eagles NOT Philadelphia."
Searching with a Web Directory
There are two main types of directories: those that are hierarchical (i.e. that lead one from a general topic to a more specific one) and those that list sources in some sort of order (most commonly alphabetical). The first type of index often contains a broad range of topics while the second usually contains sources designed to address a particular topic or concern.
Most search engines have some sort of index attached to them. More prominent and well-developed ones include The Open Directory Project, Yahoo!, and Google. Indexes are valuable for Web researchers who have an area on which they want to focus, but do not yet have a specific topic. An index can help a writer get general information or a "feel" for the topic.
- Go to Yahoo! (contains a Web directory)
- Find a topic that interests you ("education")
- Follow it through specifics ("rural education", "Rural Education Institute")
- "Rural Education Institute" is a specific topic that can be feasibly researched, either by following the listed links or by using that phrase in a keyword search.
Search Engine and Web Directory List
The following is a list of some of the most powerful search and metasearch engines and most comprehensive web directories.
- All4one: One of the first metasearch engines, All4One allows simultaneous searching of 10 major search engines.
- Bing: Microsoft's search engine.
- Dogpile: A metasearch engine that will search Google, MSN, Yahoo, and Ask.
- Environment Web Directory: A web directory that focuses on environmental and health issues.
- Excite: A search engine that lets you search by language, for video, audio, and mp3, and by relevant date.
- Google: Includes a new type of search, "Google Scholar," which allows you to search for more academically-oriented searches.
- Lycos: A search engine that allows for news searches but does not have many advanced search features.
- Metacrawler: A metasearch engine and will search other search engines.
- The Open Directory Project: One of the largest and most comprehensive human-edited directories in the world. Only higher quality websites will be listed here as each site submitted must be approved by a directory editor.
- WebCrawler: Another search engine that allows searching by location, domain name, and for multimedia.
Resources to Search the Invisible Web
The invisible web includes many types of online resources that normally cannot be found using regular search engines. The listings below can help you access these resources:
- Alexa: A website that archives older websites that are no longer available on the Internet. For example, Alexa has about 87 million websites from the 2000 election that are for the most part no longer available on the Internet.
- Complete Planet: Provides an extensive listing of databases that cannot be searched by conventional search engine technology. It provides access to lists of databases which you can then search individually.
- The Directory of Open Access Journals: Another full-text journal searchable database.
- FindArticles: Indexes over 10 million articles from a variety of different publications.
- Find Law: A comprehnsive site that provides information on legal issues organized by category.
- HighWire: Brought to you by Stanford University, HighWire press provides access to one of the largest databases of free, full-text, scholarly content.
- Infomine: A research database created by librarians for use at the university level. It includes both a browsable catalogue and searching capabilities.
- MagPortal: A search engine that will allow you to search for free online magazine articles on a wide range of topics.
Other Useful Sites for Finding Information
Other useful places to begin to search include:
- Librarians' Internet Index: Provides librarian-reviewed websites and material on a host of different topics. While this site is not exhaustive, it will provide you quality information on a large variety of topics. Some of this material is invisible-web material.
- About.com: Provides practical information on a large variety of topics written by trained professionals.
- Wikipedia: The largest free and open access encyclopedia on the internet.
- Refdesk: A site that provides reviews and a search feature for free reference materials online.
Other Strategies for Web Searching
Don't limit your Internet searching to using search engines. Be creative and think about which Internet sites might have the information you are looking for. For example, might any of the following lead you to the sites that will provide the information you are looking for?
Looking for information about job opportunities? Look at some of the sites listing job vacancies. Try university websites that sometimes list jobs through their placement offices, or try professional organizations which also sometimes list jobs in that field. Or look through the websites of various large companies because they usually have a section on job opportunities in their company.
Looking for information likely to be discussed on newsgroups or chat rooms? Look through the lists of newsgroups or use a search engine.
Looking for information about a current topic? Check the newspaper and current newsmagazine sites. Most have a search engine for articles in their publications.
Looking for data that might have been collected on a government site? Start with sites such as the Library of Congress or The White House. If the data concerns a state or a foreign country, is there a site for that political entity?