At 07:32 06-01-00 , Nancy Whittley wrote:
>I have a question.  When search engine spiders are 
>going through sites and collecting links, do they also 
>collect virtual links?  

First, how a search engine 'bot (or for that matter any 'bot) does its
thing is entirely up to that 'bot.

Whether a relative or absolute URL is used, it will be mapped to an
absolute URL before use, and is then used in the same way. Ergo, most 'bots
would presumably use it in the same way.

Strictly speculating, it is conceivable that an extremely simplistic 'bot
would peruse all relative URLs during a visit, and queue absolute URLs for
future visits. Unlikely, but conceivable.

>In other words, are you more likely to get your pages spidered more
>thoroughly by using complete links such as 
><a href="http://www.mysite.com/page.html">
>Or will they spider this type of link also:
><a href="page.html">

Unlikely to make any difference, I should think. Possible, of course, as
some search engines do use peculiar criteria. At the same time, whether the
more complex relative URLs (such as those involving use of (..)) are
resolved correctly may be problematic.

In the more subtle situations, such as servers known by multiple names, a
'bot would likely make several discrete visits before all permutations are
visited (if ever).

>I have not read anything about this..  so I was curious 
>if anyone else has ever thought about it, or maybe 
>has the answer?

If you've researched a few of the sites that claim expertise on the topic,
the lack of mention is quite suggestive. They've tried to speculate and
reverse engineer (and research documentation) on so many things I'd hardly
be surprised if one told me even what variety of beverages its developers
preferred.

I'm presuming from your example that you are referring to absolute and
relative URLs. Other things, such as links coerced via JavaScript or Java
(for example) might well be much more problematic, to the extent that I'd
suggest being quite doubtful.

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Harold A. Driscoll                 mailto:Harold(at)Driscoll.Chi.IL.US
#include <std/disclaimer>                 http://Driscoll.Chi.IL.US

• Web Designer Courses
• Off the Shelf Training Courses

HWG hwg-techniques mailing list archives, maintained by Webmasters @ IWA