Re: Errors, warnings and website rank

> I am writing in regards to the correlation or lack thereof, concerning
> errors and warnings and how websites are ranked; especially on Google. I
> understand that there are much more problematic aspects with website  
> ranking
> such as good content, proper linking techniques etc...

Google ranking is determined (fundamentally) by the number of pages on  
_other_ sites that point to pages on your site. As long as Google can  
process your site's code, it will be able to index it. And while a  
standards-compliant site is pretty much guaranteed to be "indexable",  
Google has no problem seeing through some of the more common errors  
(otherwise it would be almost useless, since less than 10% of web pages  
use completely valid mark-up).

In any case, that only influences whether a page gets correctly _indexed_  
or not, it has no influence on its Google _ranking_ (see above).

Also, the ranking will depend on what people search for. Try searching for  
"rust remover" and then "rust removal" and you'll see different results.

> yet many other websites fail miserably in your validator
> and rank at the very top of the searches.

Google and the W3C Validator are not (as far as I'm aware) connected in  
any way. Neither is Yahoo or any other search engine (different search  
engines rank pages in different ways, BTW).

> My question is why should I bother to use your
> validator at all?

I use the validator because it makes it easier to catch mistakes and  
ensure the code will be compatible with as many processors (search  
engines, browsers, etc.) - both present and future - as possible. It will  
not (and is not meant to) make a site more popular or give it a higher  
ranking in any particular search engine. For that you need to get other  
sites to link to yours, or pay for a "sponsored link", which will always  
appear at the top of the search results page.

P.S. - Google's main page doesn't validate, BTW, but I guess they're not  
too worried about not being indexed by themselves, and they prefer to save  
bandwidth by leaving out some required elements.

RMN
~~~

Received on Thursday, 20 November 2008 08:32:47 UTC