| Author |
Message |
Anonymous
|
Posted: 24 Apr 2006 20:09:56
Skalinks finds many times invalid links although our site was sometimes correctly published on the remote site.......(?)
My setting is to find precise backlink addresses.
Our site being http://www.red4web.com/
For example please check those 2 >>
this link
http://www.freelancedesigners.com/html/link.cfm?page=www.red4web.com/d irectory/0techno/linking/index.html
that can be found at
http://fuengirola_esp.webdesign.freelancedesigners.com/
does not find it although the link exists
or
in another case:
LinkBack URL for "Surfwebtips is a human edited free submission web directory" is INVALID! Please, check it: LinkBack URL - http://www.surfwebtips.com/servlet/internet?resultatcat=Internet&valtA ntal=10
Alternative domain: Surfwebtips is a human edited free submission web directory
The question is:
1. what does it search for when the script checks a remote site, does it search for "http://www.mysite.com"?
2. is the backlink in the first example valid for Google as pointing to our site? and how can I check it? They are a professional directory, are they playing tricks by obtaining our pure link and giving back a dead link from Google's point of view?
3. In the second case the link is correct on the remote site but the skalinks program does not find it. Why? And how to get Skalinks to accept it?
Many questions but I think they may be useful to others, and I have the impression that I will have lots of links like that in the future, does it mean that many links may not be valid, too complex and finally of no use in terms of Google's or Alexa's ranking.
I really wonder what you will answer to those..
Thanks for helping with this, and for the very elegant program that you created: I'm impressed otherwise.
/Frank
|
skalinks
Admin
|
Posted: 25 Apr 2006 06:56:20
Hello Frank
Script works fine.
In first, what is the Alternative domain?
This is specially for 3 way link exchange, and you whould specify field Alternative domain for link as otherdomain.com.
In second, how skalinks detect linkback URL for link:
1) If the field Alternative domain is specified, then skalinks try to link following to this domain on LinkBack URL, else, skalinks try to find link following to domain, where skalinks is situated. Therefore, if you do not need 3 way link exchange, you do not need to specify Alternative Domain for link, and in this way SkaLinks will try to find backlink with your domain.
2) Detection of LinkBack valid makes by special system, which uses Regular Expression , this is a special feature to find text or part of text in text.
Let see your example,
You said that http://fuengirola_esp.webdesign.freelancedesigners.com/ has backlink to your site ( domain: red4web.com ), and Skalinks wrote that this backlink is invalid, right? Let's check this backlink manually. If we open
http://fuengirola_esp.webdesign.freelancedesigners.com/ in browser and try to find link wiith title red4web, we find it, BUT, open the source of this page ( in IE Menu->View->Source, in Mozilla CTRL+U ), and try to find domain red4web, and you will find this
<a href="http://www.freelancedesigners.com/html/link.cfm?page=www.red4web .com/directory/0techno/linking/index.html" rel="nofollow" class="linked_listing_display_name" target=listing>
Red4Web.com Web Design</a>
THERE ARE 2 SERIOUS MISTAKES.
First does you see the domain name of this link http://www.freelancedesigners.com - this is not your domain, and google will not see your domain in this link.
Second, this link has attribute rel="nofollow" - this is SPECIAL tag for search engine, which do not allow search engine follow to this link.
Draw a conclusion by yourself :-)
I do not know why the owners of this site do this, may be this their policy or something else.
Also, LinkBack URL verification system works in 2 mode:
Simple mode - just search valid link to your site.
Full mode - search valid backlink and try to go to found link ( need more time and resources, but more efective )
Did I answered to your question?
Thank you for choise SkaLinks!
|
koolred
Member
|
Posted: 25 Apr 2006 18:09:10
Hi,
Yes I never thought really that Skalinks is doing the work wrong. I could spot that the second link was not okay but I should have looked at the source too. Thanks for spotting this one and for taking time to tell more about your program, it's a very good piece of knowledge for everyone.
In the other very interesting (3 way link) case (http://www.surfwebtips.com/servlet/internet?resultatcat=Internet&valt Antal=10), I think I understand your explanation only partly. No actually I was not looking for a 3 way linking. Should I change this link in the "LinkbackURL" text field in order to enable it to accept the existing backlink on this page.
Because finally what imports to me is that the link exists and I want to have skalinks to accept it, from a selfish user's point of view, otherwise those kind of links accumulate in the not valid backlinks although it's not exactly true in this case since it is valid to Google, that's enough for me.
Thanks, that was the best and most documented answer I ever got in a forum, I really appreciate, THANKS.
/Frank
|
skalinks
Admin
|
Posted: 26 Apr 2006 06:50:23
Hi koolred
Tag rel="nofollow" was inmplemented by Google at first, and Google ignore all links with this tag. Also, if you still wish that Skalinks lets pass this links, no problem, we can dscribe you how you can do it, or we can do it for you.
Thank you Frank
|
Anonymous
|
Posted: 26 Apr 2006 18:13:19
Sorry this email was confusing, I did not mean this link, I meant this link:
http://www.surfwebtips.com/servlet/internet?resultatcat=Internet&valtA ntal=10
how to let it pass the test although it is a special case of a 3 way link?
Thanks
|
Anonymous
|
Posted: 26 Apr 2006 18:16:34
error in the link above:
http://www.surfwebtips.com/servlet/internet?resultatcat=Internet&valtA ntal=10
|
skalinks
Admin
|
Posted: 1 May 2006 01:32:38
Hello,
Open SkaLinks.class.php file in SkaLinks_include folder
Find this code:
preg_match( "/<a\s([^<>]*?\s)*?href\s*?=\s*\"\s*(http:\/\/(w{3}\.)?{$matches[3]}\/ ?(.*?)*?)\"([^<>]*?\s*)*?>.+?<\/\s*a\s*>/is", $body_post, $match );
if ( !$match[0] )
{
preg_match( "/<a\s([^<>]*?\s)*?href\s*?=\s*\'\s*(http:\/\/(w{3}\.)?{$matches[3]}\/ ?(.*?)*?)\'([^<>]*?\s*)*?>.+?<\/\s*a\s*>/is", $body_post, $match );
}
if ( !$match[0] )
{
preg_match( "/<a\s([^<>]*?\s)*?href\s*?=\s*(http:\/\/(w{3}\.)?{$matches[3]}\/?(.*? )*?)(\s*[^<>]*?\s*)*?>.+?<\/\s*a\s*>/is", $body_post, $match );
}
print_r($match);
if ( $match[0] && ( !preg_match( "/rel\s*=\s*(nofollow|'nofollow'|\"nofollow\")/ims", $match[0] ) ) && !substr_count( $match[0], "<!" ) )
{
And replace it with:
preg_match( "/{$matches[3]}/is", $body_post, $match );
if ( $match[0] && !substr_count( $match[0], "<!" ) )
{
|
|
