Absolute vs relative links and Google PR
Christos
know if it matters whether or not a site uses absolute or relative
links in determining google page rankings? I would think using
absolute links couldn't hurt, but if there's no need to change our
site (currently uses relative links), then why bother.
I'm curious because while google has crawled our site several times,
only the main page and our promotion page have been included in the
google database. The other 10 or so pages, which I assumed would be
crawled automatically, are nowhere to be found.
Thanks.
-----------------------------------
Christos Georgoulas
Casino Strategy Cards
Play Smart...Win More
chri...@casinostrategycards.com
http://www.CasinoStrategyCards.com
-----------------------------------
Gateway Farm Alpacas
"Christos" <chri...@casinostrategycards.com> wrote in message
news:24f3272f.04030...@posting.google.com...
> I've done some searches but cannot find a sure answer. Does anyone
> know if it matters whether or not a site uses absolute or relative
> links in determining google page rankings? I would think using
> absolute links couldn't hurt, but if there's no need to change our
> site (currently uses relative links), then why bother.
>
> I'm curious because while google has crawled our site several times,
> only the main page and our promotion page have been included in the
> google database. The other 10 or so pages, which I assumed would be
> crawled automatically, are nowhere to be found.
>
> Thanks.
>
I can only speak anecdotally, but my rankings improved when I went to
absolute URLs.
Google is not always good at resolving different URLs pointing to the same
page, and then the PR suffers. The relevancy of this is that if someone
deep links to your site using an absolute URL, but you are using relative
URLs, Google may not see them as the same page. In that case, PR will not
transfer as expected, and worse, you may be penalized for duplicate content.
--
John Merrell
Gateway Farm Alpacas
www.gateway-alpacas.com
Philipp Lenssen
> I've done some searches but cannot find a sure answer. Does anyone
> know if it matters whether or not a site uses absolute or relative
> links in determining google page rankings? I would think using
> absolute links couldn't hurt, but if there's no need to change our
> site (currently uses relative links), then why bother.
>
> I'm curious because while google has crawled our site several times,
> only the main page and our promotion page have been included in the
> google database. The other 10 or so pages, which I assumed would be
> crawled automatically, are nowhere to be found.
>
I believe it makes absolutely no difference. Use "relative absolute"
links like this: /tutorial/bla.html or / or /tutorial/ and so on and
you should be safe. I have no problems with this. I don't however
normally use ../bla/ and so on so you might have problems with that. In
any case differentiate between /bla and /bla/ as those are two
different URLs. Also, www.example.com or just example.com make a
difference. Though not a big difference, Google seems to get better at
resolving this. But technically those are 2 different URLs.
--
Google Blogoscoped
http://blog.outer-court.com
Ringo Li
> page, and then the PR suffers. The relevancy of this is that if someone
> deep links to your site using an absolute URL, but you are using relative
> URLs, Google may not see them as the same page. In that case, PR will not
> transfer as expected, and worse, you may be penalized for duplicate content.
There is no such penalty. PR is assigned to each URL, whatever way
the other sites point to you site which you have no control. You can
have a different PR for:
http://www.yourdomain.com
http://yourdomain.com
yourdomain.com
http://www.yourdomain.com/index.htm
http://yourdomain.com/index.htm
yourdomain.com/index.htm
You can have six different PR just like this. There is no penalty for
duplication because it is not duplication. It does not matter Google
"cannot resolve" that it may be the same page. Some of this URL may
have a PR0 which other URL may have a PR5. It is entirely normal.
Gateway Farm Alpacas
"Ringo Li" <ato...@hotmail.com> wrote in message
news:3920d2ba.04031...@posting.google.com...
>
> There is no such penalty. PR is assigned to each URL, whatever way
> the other sites point to you site which you have no control.
Google itself states (http://www.google.com/webmasters/guidelines.html)
1. Avoid hidden text or hidden links.
2. Don't employ cloaking or sneaky redirects.
3. Don't send automated queries to Google.
4. Don't load pages with irrelevant words.
5. Don't create multiple pages, subdomains, or domains with substantially
duplicate content.
6. Avoid "doorway" pages created just for search engines, or other "cookie
cutter" approaches such as affiliate programs with little or no original
content.
Number five is the relevent entry here. While not stating directly that
there is a penalty for duplicate content, many have provided evidence that
one does in fact exist.
> You canhave a different PR for:
>
> http://www.yourdomain.com
> http://yourdomain.com
> yourdomain.com
> http://www.yourdomain.com/index.htm
> http://yourdomain.com/index.htm
> yourdomain.com/index.htm
>
> You can have six different PR just like this. There is no penalty for
> duplication because it is not duplication. It does not matter Google
> "cannot resolve" that it may be the same page. Some of this URL may
> have a PR0 which other URL may have a PR5. It is entirely normal.
Help me to understand how it can "not matter" that the same page carries
varying amounts of PR according to the URL it is called by.
Your analysis of six different ways to call a page is close, although you
forgot variations like 'default' et.al. It misses the point though. The
original question was, "Does anyone know if it matters whether or not a site
uses absolute or relative links in determining google page rankings?"
Even by your analysis, the answer is "Yes".
The simple fact is that inbound links will not use relative URLs.
Therefore, common sense supports the idea that absolute URLs on the site
will most likely match inbound links and insure that PR is not diluted
amongst what Google sees as a bunch of different pages.
Additionally, it is often a good idea to use some generic rules in .htaccess
to permanently redirect the common variations to one page. (Unfortunately,
Y, Ink, and Ask don't seem to know what a 301 redirect is).
Jay
news:c2mo7p$1ulc43$1...@ID-203055.news.uni-berlin.de...
Isn't the base reference in the http response headers? I would think it
would be fairly easy for google to resove relative urls.
Jay
www.seo-highrankings.com
news:U504c.15133$oP.1938@lakeread03...
I don't believe relative links are counted by Google when they are going
back to the homepage. For instance, when setting up a link from an internal
page in your site back to teh index page you can do it two ways;
http://www.yoururl.com/nidex.htm
or
index.htm
I do not believe that Google counts the second one as a fully qualified link
so that it passes page rank.
--
James Taylor
http://www.AICompany.com - SEO, Web Development and Hosting
http://www.SEO-highrankings.com -FREE SEO TOOLS
Philipp Lenssen
>
> I don't believe relative links are counted by Google when they are
> going back to the homepage. For instance, when setting up a link from
> an internal page in your site back to teh index page you can do it
> two ways;
>
> http://www.yoururl.com/nidex.htm
> or
> index.htm
>
> I do not believe that Google counts the second one as a fully
> qualified link so that it passes page rank.
And you belief is based on ...?
James Taylor
Over seven years and 100 sites that we have SEO'd and all have made it into
#1 at Google for their terms. :-)
I have never seen an example of a site where a relative link gets counted
towards PR.
If you have one, I would be happy to take a look.
--
James Taylor
http://www.aicompany.com
http://www.seo-highrankings.com
James Taylor
"John A." <no....@spammers.virginiaquilter.allowed.com> wrote in message
news:9bp150d1mn8l2rreo...@4ax.com...
Please don't be confused between your opinion and real data.
If you can find one site that shares PR throughout the entire site, every
page with relative URL linking back to the home page, I would like to see
it.
I just now tried vacuum cleaner and several used cars but to be honest, have
no time to spend on opinions without substantiation...got a business to run.
In this NG we try to present ideas and information built upon things we can
demonstrate...otherwise it is pretty much worthless.
I have not seen a single site in seven years that can do this.
If you have just one, I will take a look.
Christos
> back to the homepage. For instance, when setting up a link from an internal
> page in your site back to teh index page you can do it two ways;
>
> http://www.yoururl.com/nidex.htm
> or
> index.htm
>
> I do not believe that Google counts the second one as a fully qualified link
> so that it passes page rank.
Thank you all for the help.
From what's been said, it seems that the relative links back to my
homepage which are now just "default.shtml", would be better served,
if I made them all "http://www.casinostrategycards.com". At the very
least, it couldn't hurt, and at the best, it may help.
www.seo-highrankings.com
> On Thu, 11 Mar 2004 14:27:30 -0800, "James Taylor" <ja...@nothere.com>
>
> I was checking for new Mozilla themes and just for kicks searched one
> of the pages' source for a "/" link, and there it was. No links to the
> root with the hostname, too.
>
> I checked for backlinks to the root page...
>
>
http://www2.google.com/search?q=link:themes.mozdev.org&num=100&hl=en&lr=&ie=UTF-8&c2coff=1&safe=off&filter=0
>
> ...and found several pages within the site recognized by Google as
> linking to the root. I checked the first one listed (about 2/3 down
> the 1st 100 backlinks)...
>
> http://themes.mozdev.org/enhancements.html
>
> ...and found the same thing: all links to the root coded as "/".
>
> JA
John,
Dude...you are a bit confused.
This is a relative link;
<A HREF="index.htm">keyword</A>
This is not;
<A HREF="http://www.yourdomainname.com/index.htm">keyword</A>
The examples you provided in you rprevious post in fact proe my point in
that if you take the time to LOOK you will see their homepage link is in
fact fully qualified...meaning that they reference
http://theirdomainname.com/
Next time, B4 you get snippy, be sure you undrstand what you are talking
about.
www.seo-highrankings.com
Hi Christos,
Yes,
Google will not count the relative link as able to pass pagerank.
Fully qualify you rlink anhd you shoul be set to go.
Be sure to give Google a few weeks or more to spider it all in and make the
update.
www.seo-highrankings.com
>
>
> >The examples you provided in you rprevious post in fact proe my point in
> >that if you take the time to LOOK you will see their homepage link is in
> >fact fully qualified...meaning that they reference
> >http://theirdomainname.com/
>
> status bar? In the source it's "/". The status bar shows the parsed
> URL.
>
> >Next time, B4 you get snippy, be sure you undrstand what you are talking
> >about.
>
>
> The only problem I've ever encountered with relative links was with
> Inktomi being unable to parse "../../../filename" when it was in a
> directory less than two dirs deep. (The coding was a side effect of
> the local directory tree in which editing was being done.) No other
> SEs or browsers had a problem with it, but Inktomi choked. After a
> while they apparently fixed it, though. Google's never given me any
> problems. There was one Japanese SE (goo?) that seemed to get confused
> occasionally about which domain it was trying to spider.
>
> But getting back to links to site index pages and PR, I stand by my
> other statement that any difference between the way Google handles one
> and the other ("/" vs "index.htm" OR "/index.htm", etc.) has nothing
> to do with relative vs. absolute linking. It has to do with whether
> Google recognizes whatever the link parses to as being the same as
> "http://www.yourdomain.com/"
>
> Common index file names like index.htm and index.html seem to be
> recognized. I think there are others as well. It's when you wander
> away from the names that Google recognizes that you seem to get into
> trouble.
>
> There may be some analysis going on to determine if they are actually
> the same on the site in question. As I posted in another reply in this
> thread, I checked backlinks for "/", "/index.html", and "/index.htm"
> on my site and they were all the same. "index.asp", "default.asp", and
> "index.php" were not recognized as being equivalent.
> www.microsoft.com/sql/default.asp had identical backlinks to
> www.microsoft.com/sql/, www.microsoft.com/sql/index.html, and
> www.microsoft.com/sql/index.htm though.
>
> IIRC, Google's recognition of index files was discussed in another
> thread here some time ago.
>
> JA
I believe we are still talking about two different things here.
Take a close look at the site you mentioned;
http://themes.mozdev.org/enhancements.html
Now type in LINK:http://themes.mozdev.org/enhancements.html into Google.
See how they are NOT counting the links back to the homepage? Because they
are "relative".
Now, type in this in Google;
link:www.Microsoft.com
See how those fully qualified links are all counted as having PR and pass it
along?
It has nothing to do with slashes (/). Let's forget about slashes because
they are irrelevant to this conversation and have nothing to do with fully
qualified links in the abstract...but if they did, the slashes are just fine
in the URL, as evidenced by my Microsoft example.
www.seo-highrankings.com
Also,
You cannot count a subdomain in the "link:" request because in your MX
Record of your website, a subdomain is treated similar to a different
domain...which is in fact how Google treats it as well.
So, in order for the results to show a "relative" link back, it would have
to appear as "realindex.html" after you type in the
"link:http://iconpacks.mozdev.org/realindex.html"
You will only see subdomains, which are not considered "relative" in that
they have a fully qualified pointer at the MX Record.
www.seo-serps.com
<james....@aicNOomSPpaAMny.com> wrote:
>Take a close look at the site you mentioned;
>http://themes.mozdev.org/enhancements.html
>
>Now type in LINK:http://themes.mozdev.org/enhancements.html into Google.
>
>See how they are NOT counting the links back to the homepage? Because they
>are "relative".
Hi James,
You have made an error in the above. The backlink check should go to
the home page (LINK:http://themes.mozdev.org) not the linking page as
you've indicated above.
You are saying relative links such as this one-
<a class="first-item" href="/">Home</a>
As found on this page-
http://themes.mozdev.org/enhancements.html
Doesn't count as a backlink in google to the home page because
relative links aren't treated as backlinks.
http://themes.mozdev.org (this being the home page)
Is that what you are saying?
If so take a look at the backlinks for http://themes.mozdev.org with
the search below-
You'll find this page http://themes.mozdev.org/enhancements.html is
listed as a backlink.
Since this link <a class="first-item" href="/">Home</a> is the only
link to the home page of this sub domain it proves Google does treat
relative links as backlinks.
There are potential SEO benefits to using both relative and absolute
links, all depends on the domain and the situation.
David
_
http://www.seo-serps.com
www.seo-highrankings.com
I believe Google treats subdomains separatly from their domains right?
I see no evidence of the root domain relative link being counted.
Do you see it and I am missing it?
--
James Taylor
http://www.AICompany.com - SEO, Web Development and Hosting
http://www.SEO-highrankings.com -FREE SEO TOOLS
"www.seo-serps.com" <ooar...@AMntlworld.com> wrote in message
news:mje250t65egf4hhnj...@4ax.com...
www.seo-highrankings.com
If I have erred, I apologize, but I still don't see evidence of the root
domain relative link being counted.
www.seo-highrankings.com
> On Thu, 11 Mar 2004 20:16:58 -0800, "www.seo-highrankings.com"
> <james....@aicNOomSPpaAMny.com> wrote:
>
> >BIG OLE SNIP
> >
> >Also,
> >
> >You cannot count a subdomain in the "link:" request because in your MX
> >Record of your website, a subdomain is treated similar to a different
> >domain...which is in fact how Google treats it as well.
>
> what I claimed at all. Remember, I said the backlinks in question (the
> ones from the same host name) were about 2/3 of the way down the first
> page (if 100 per page.)
>
> >So, in order for the results to show a "relative" link back, it would
have
> >to appear as "realindex.html" after you type in the
> >"link:http://iconpacks.mozdev.org/realindex.html"
>
> link code. It's showing where the links are coming from anyway, not
> where they're going to.
>
> >You will only see subdomains, which are not considered "relative" in that
> >they have a fully qualified pointer at the MX Record.
>
> than "iconpacks.mozdev.org/", "iconpacks.mozdev.org/index.htm", and
> "iconpacks.mozdev.org/index.html", all of which have the exact same
> list since they are considered by Google to all be the same page. It
> just illustrates my point about Google recognizing some file names as
> index files and not others.
>
> This would be so much easier to resolve if only Google let us search
> for snippets in page source code. :)
>
> JA
Agreed.
I could be wrong on this. You have brought up some good points.
I have been using Google's results as examples but they may not be showing
the correct relative URL.
www.seo-serps.com
<james....@aicNOomSPpaAMny.com> wrote:
>Hi David,
>
>I believe Google treats subdomains separatly from their domains right?
Ahh, I see where you are going wrong. Google does treat sub domains as
a separate entity (which it is really), therefore this link-
<a class="first-item" href="/">Home</a>
Isn't pointing at the main domain (http://mozdev.org), it's pointing
at the sub domain http://themes.mozdev.org
Google sees it this way and so do browsers. If you click the home link
on the internal pages it takes you to the home page of the sub domain
not the main domain.
>I see no evidence of the root domain relative link being counted.
So it's counting as a link to the home page of the sub domain which as
far as Google (and browsers) is concerned is a domain in it's own
right.
Since sub domains are treated as a separate domain you have to use
absolute links to transfer PR etc... to the main domain or other sub
domains. Treat sub domains like you would any other domain.
The important thing is relative links do count. If this was a relative
link on a standard domain (not a sub domain) it would take you to the
home page in a browser and Google will treat it as a backlink to the
main domain.
I mix my relative and absolute links all the time since sometimes they
help with the SEO process other times they don't. I do know from
experience relative links are treated as backlinks to any page they
link to (as they should).
I don't like using slashes though. It's not for a good reason, just
don't like them. This is why when it comes to home page links I tend
to use absolute. Not always, sometimes I'll use index.html or
index.asp.
BTW meant to email you the last post, clicked the wrong button and
sent it to the NG.
David
_
http://www.seo-serps.com