Hi All

I was doing some messing about on Google last night doing:

link:www.whatever.com

To see how many inbound links some of my competitors had.

Any road up, I noticed that some sites had links back to themselves that
were being counted. This had been done by, for example, linking the home
button to http://www.whatever.com instead of just linking to index.html.

Does the panel have any thought on whether this will improve page rank? I
know it's more complicated that this but seeing some of the sites that rank
higher than mine I was scratching my head looking for other reasons.

Andy

On Fri, 27 Jan 2006, Andy Jacobs wrote:

Linking to "index.html" and thus creating two different URLs for
the same resource is bad style anyway - whether it affects search
engines or not. If you understand German, you might want to read
http://schneegans.de/web/kanonische-adressen/

Google has quite a lot of hits for "index.html":
http://www.google.com/search?q=index.html
http://www.google.com/search?q=index.htm
revealing the careless webmasters.

--
Netscape 3.04 does everything I need, and it's utterly reliable.
Why should I switch? Peter T. Daniels in <news:sci.lang>

Andy Jacobs wrote:

You're far better off using Yahoo. Google deliberately restrict the
results from a link: search, so you will get a very limited set of
pages, whereas Yahoo gives you the lot.

I'm not really sure what you mean - some pages had links to
themselves? I don't see that as news...

I don't suppose that is done to deliberately manipulate search engines
- it can be useful for people who save copies of pages onto their
computer and then forget where they are from - if there is an absolute
link back to the homepage it will enable them to locate it.

I suspect it will have very little effect. I was reading some stuff
about Pagerank earlier this week (I think it was on www.sitepoint.com
but I'm not too sure), that basically said it's not worth worrying
about. Google doesn't just give a page rank to a page based on how
many inbound links it has (although it did in its early days), but
rather, it assess the value of each inbound link *for the search
phrase in question*.

I can't believe it would be fooled by intra-site links.

--
Stevie D
\\\\\ ///// Bringing dating agencies to the
\\\\\\\__X__/////// common hedgehog since 2001 - "HedgeHugs"
___\\\\\\\'/ \'///////_____________________________________________

On 27/1/06 5:05 pm, in article
Pine.GSO.4.44.0601271759020.9680-100...ni-hannover.de,
"Andreas Prilop" <nhtcapri@rrzn-user.uni-hannover.de> wrote:

So should 'Home' links ALWAYS link to http://www.whatever.com then?

Andy

On Fri, 27 Jan 2006, Andy Jacobs wrote:

In principle, yes.
Depending on your concrete situation, you might also write
<a href="./">
or
<a href="/">

--
Netscape 3.04 does everything I need, and it's utterly reliable.
Why should I switch? Peter T. Daniels in <news:sci.lang>

On Fri, 27 Jan 2006 17:10:35 +0000, Andy Jacobs wrote:

One can, and should, go a lot further than this. The general idea is that
the implementation -- the technology if you like -- of a resource should
be hidden as far as is possible because (a) it is of no concern to the
reader, and (b) it makes for more flexible designs if you do so.

The fact that the "index" page for a directory is called index.html (or
default.html or home.html or whatever) is an implementation matter and I
would prefer to link like href="reports" rather than
href="reports/default.html".

You get even more flexibility if you hide the extension since that is also
an implemenation matter. I prefer to have a link to "reports/2006" rather
than "reports/2006.html". If I choose to switch to a PHP implemenation,
I don't need to change any links and no bookmarks get invalidated.

Obviously you need the cooperation of the server, but can't imagine there
are any servers (other than toy ones) that would not cooperate.

--
Ben.