Web-Design and Algorithms

While looking for near-magical formulae for the perfect placement and counts of keywords, many webmasters forget to allow for the effects that web-design issues such as the use of javascript, tables or images, may have upon the spider's perception.

A search engine spider is not a browser. It does not 'see' a page in the way that a browser does. In fact a spider is much more like a word-processor or text editor. It looks at the text, performs complex search and replace operations, and then attempts to process a page, similar to the way that a word processor might attempt to check grammar.

Anything that is not actual text is almost akin to an obstacle for the spider. It is simply extra characters in the text it sees that it has to decide what to do with. This page, for example, uses tables and graphics to present information more clearly for the human eye.

This section of Rithmix will look at various basic web-design issues and see if and how these affect the algorithms.

Remember above all that these pages are primarily a tutorial in how to design your own search engine algorithm analysis pages and experiments. These pages here are very public and therefore could be subject to tampering or abuse which would affect the results.

The following are the pages in this test:

1. web-design_1.html - This is the plainest possible HTML page consisting of absolutely minimal tags and with the emphasis on pure text content.

2. web-design_2.html - This page places the text of the above inside a table layout with nested tables. This adds a lot to the layout but leaves the content unchanged.

3. web-design_3.html - This page adds some graphics to the page to improve the appearance and design of the plain page. We replace one incidence of the keyword 'Rithmix' with an image which has the keyword in the ALT attribute

4. web-design_4.html - This page adds some Javascript functions to the content while again leaving as much of the content itself untouched as possible.

5. web-design_5.html - This page is an exact duplicate of page 1.

6. web-design_6.html - This page is an exact duplicate of page 2.

7. web-design_7.html - this page is an exact duplicate of page 3.

8. web-design_8.html - This page is an exact duplicate of page 4.

The reason for having a duplicate of each page is to isolate rankings that are based on content, from other factors. If the layout is the ranking factor then we should see the identical pages ranked together. For Example: If page 1 were the highest ranked, we would expect to see page 5 (the duplicate) right behind it if the layout is indeed the factor.