The first and most common type of search tool is the 'crawler-based search engine'. Crawler-based search engines, such as Google, Inktomi and Alta Vista, create their database of site listings by 'crawling' or 'spidering' the web. They catalogue the web pages that they find, then each search engine uses it's own sophisticated algorithms and rules to determine which pages will be listed and in what order when you search for a particular word or phrase. In most cases also, on a regular basis, the search engine's 'spider' or 'crawler' will revisit pages that are already in its database in order to ensure that its catalogued snapshot of each page is as current as possible.

Unlike search engines, "human-edited directories" are created and edited by human editors, rather than spiders or crawlers. A human editor is responsible for the hierarchy and categorisation of each site that is listed in the database. Because of the human role in reviewing and assessing the relevance and quality of each site before inclusion in the database, some directories may actually provide better results than search engines. Dmoz.org and Looksmart.com are two of the more popular human-edited directories.

'Hybrid search engines' offer a 'best of both worlds' scenario. They offer combined or mixed results from a human-edited directory, as well as results from a regular crawler-based search engine. Hybrids usually favour the results of one of its sources of results over the other, so for example, at MSN-Search, a search is likely to present Looksmart.com's human-edited results first followed by Inktomi's crawler results.

'Meta search engines' do not have their own crawlers or spiders, nor do they have their own database. Instead they allow surfers to search the databases of several engines simultaneously, with the search results from these various engines being combined on the same page(s) of results. So for example, a search at Ixquick.com will see results from a number of engines including All The Web, Alta Vista and Teoma, all on the same page. 'Pay Per Click Engines' such as Overture and FindWhat have a very simple formula. Companies or web sites ask to be listed in their database, and they pay the search engine a certain amount every time their site's listing is clicked.

So where does this leave the ever-popular Yahoo? Up to a few weeks ago, Yahoo could be accurately described as a hybrid search engine. Back then, it pulled results from its own human-edited directory first, and the rest came directly from Google's database. Now, Yahoo's results are being pulled directly from the Google database, effectively making it a mirror of Google.

http://www.jamaica-gleaner.com/gleaner/20021120/business/business4.html