Improving your search-engine status - Part 1
This article is about improving your search-engine ratings. But before
venturing into the technical territory, let us first set the scene. So,
imagine that...
.. relaxing in the bath after a hard day's work, you discover to your horror that your cherished 'yellow duck sponge' has been thrown out. Even worse - the special place in the cupboard reserved for such sponges is entirely bare.
In the past, such a domestic tragedy may have necessitated a trip into town to restock the sponge menagerie. But in this Internet-enabled age you simply reach over to your customised waterproof laptop, fire up a browser, and navigate to your search-engine of choice. Within seconds you have a list of yellow duck sponge manufactures to peruse at your leisure...
Unfortunately, the least plausible part of this story is the last sentence. For instance, when we tried the search phrase 'yellow duck sponge' on Google, it returned 4780 documents supposedly matching the given criteria. But the first of these described a young artist's success in cutting a sponge into a duck shape, and amongst the others were 'Vulcan Software Cheats and Solutions' and 'Whimsical Teapots'. Altogether, there were perhaps only two links on the first page of results that seemed worthwhile pursuing.
Now, as a seeker of novelty bath items we could perhaps have tried a number of different strategies - refining our search, for instance, or trawling through the pages of results. But research suggests that non-expert users of the Internet - whom we can assume to be in the majority - just aren't good at reformulating search terms (see 'What's Wrong with Internet Searching',
http://www.dlib.org/dlib/march97/bt/03pollock.html). And hardly anyone ever goes past the first couple of pages of search results, whether expert users or not.
The onus, therefore, is on the manufacturers of yellow sponge ducks - as well as purveyors of more mundane wares - to make sure that their search results come high on the search results of the average searcher.
When you submit a query to a search-engine, it delves into its big bag of information (a technical term) to come up with suggestions. The information it uses to order these suggestions derives from four general sources (although not every search-engine uses every source).
A search-engine's software can automatically load up and navigate through websites, indexing what it finds there. It is common to describe this process in B-movie terms, as visits by 'a crawler', 'a spider', or 'a robot'.
Probably the biggest repository of such information is Inktomi, which sells its services directly to other search engines.
Sometimes search results are influenced by the reports of people who have looked at sites and decided how to catalogue them.
The best known user of human analysis is Yahoo, which employs some 150 editors. The decisions of these editors results in Yahoo's directory structure, and the sites featured in the directory predominate in search results. Also producing a directory structure, which is utilised by various search engines, is the 'Open Directory' project, which takes data from volunteers around the world.
The direct results of 'Ask Jeeves' are also moderated by reviewers, who check on each site's suitability to provide authoritative answers to particular questions.
Search-engines may gather two sorts of information relating to site popularity. The first is the number of other sites which link through to each page. The second is the number of times a link is clicked on when it is presented in search results.
The heaviest user of the first index of popularity is Google. In fact, if you don't have a number of people linking to you, it's likely that you won't appear on Google at all.
The second index of popularity is measured by Direct Hit, which looks at how people use the search results returned by its partners. These partners - who include AltaVista - then feed back the results into subsequent search results. Of course, to become popular in this way, you do have to appear on the searches in the first place.
A very few search-engines rank sites based upon the size of the cheque provided to them by the site's owners. The prime example of this is Goto, which grew out of the Web Wide Worm.
next part
|
